From alex at alexfrancis.org.uk  Tue Jul  1 09:13:31 2008
From: alex at alexfrancis.org.uk (Alex Francis)
Date: Tue, 1 Jul 2008 09:13:31 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
	things from a different angle...
In-Reply-To: <48695161.1030201@cosmicperl.com>
References: <48695161.1030201@cosmicperl.com>
Message-ID: <16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>

On Mon, Jun 30, 2008 at 10:34 PM, Lyle <webmaster at cosmicperl.com> wrote:
> Hi All,
>  These days I hear a lot of:-
> "You should use language X for this, language Y for that and language Z
> for the other"
> "Because language X was designed for this, Y is most commonly used for
> that, and Z was made for that OS"
>

Hi Lyle

While I'm (of course) very much in favour of Perl, I think for the
most part it's possible to hire bright programmers who can adapt to
whatever language or platform they need to. I've found the bigger
challenge to be in helping non-technical senior management to make
sensible choices and in helping them to trust their expensive
technical staff.

I've worked places where the language/platform selection has been
based on some made-up nonsense about customers being more comfortable
with X or Y and if we use Z then we won't be able to sell (Z being
anything that isn't Java or .net). I've also seen what should have
been considered strategic decisions become panic-stricken knee-jerk
changes which waste huge amounts of money.

There's also a problem in between the two - convincing the
non-technical manager that it's less risky to convert your existing
team from Z to X than it is to try and hire new staff with 5 years
experience in X. This can be a distinct problem with Perl to be
honest: if you find yourself having to look for an experienced Perl
programmer, well, good luck.

All good fun!

Everywhere I've worked has had legacy code in a language that the
company doesn't write in any more (often C, but you'd be amazed by
some of the stuff I saw at the county council I worked for:-). So
that's just a fact of life for the most part.

Something I would say about Perl is that as a programmer it exposes
you to a lot of concepts that make other languages easier to pick up
(when you inevitably have to pick them up). Perl's closures made a
"meta-programming in javascript" talk I saw make perfect sense and
some of Ruby's features were familiar too. Perl's references mean I'm
not terrified of pointers. Perl's OO (if you can call it that) made
Java an easy (if tedious) switch (and Java made C# straigtforward to
learn). Perl's regexes meant I could help my Dad with his PHP
questions.

Alex

From pmh at edison.ioppublishing.com  Tue Jul  1 09:19:13 2008
From: pmh at edison.ioppublishing.com (Peter Haworth)
Date: Tue,  1 Jul 2008 09:19:13 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
References: <4868F684.8030901@cosmicperl.com>
	<50fec4060806300834o35015e21ke72623f54385ad78@mail.gmail.com>
	<50fec4060806300953paa2c24ax14aaa651014fe356@mail.gmail.com>
	<48694664.4080800@cosmicperl.com>
Message-ID: <PGM.20080701081913.7623.9084@edison.ioppublishing.com>

An embedded and charset-unspecified text was scrubbed...
Name: not available
Url: http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080701/d26a3317/attachment.ksh 

From mjr at phonecoop.coop  Tue Jul  1 11:50:52 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Tue, 01 Jul 2008 11:50:52 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <4868F684.8030901@cosmicperl.com>
References: <4868F684.8030901@cosmicperl.com>
Message-ID: <486a0c0c.qe86ZIM3btvBshhc%mjr@phonecoop.coop>

Lyle <webmaster at cosmicperl.com> wrote:
> http://perl.bristolbath.org/tshirt/front.png
> http://perl.bristolbath.org/tshirt/back.png
>
> I'm not sure on the font, or about having so many logos on the back. I 
> am happy with the slogan and the duck logo :)
>
> Comments and suggestions!

About the back: I already have an O'Reilly t-shirt that I don't wear
much (because I don't like them).  Why would I want another one? 
Much better to make the Powered By Perl logo the main one, IMO.

Please can the eventual t-shirt be at least meet the ETI Base Code, or
preferably be Fair Trade or Cooperatively-produced?  I want to avoid
buying clothes produced with unfair labour.

Thanks,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From mjr at phonecoop.coop  Tue Jul  1 12:20:55 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Tue, 01 Jul 2008 12:20:55 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <4868F684.8030901@cosmicperl.com>
References: <4868F684.8030901@cosmicperl.com>
Message-ID: <486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>

Lyle <webmaster at cosmicperl.com> wrote:
> http://perl.bristolbath.org/tshirt/front.png
> http://perl.bristolbath.org/tshirt/back.png

Actually, this just came up over on http://lists.debian.org/debian-perl/

That t-shirt back may be infringing the Perl Foundation's trademark.
See http://www.perlfoundation.org/perl_trademark for their claims.

The camel claims are http://www.oreillynet.com/pub/a/oreilly/perl/usage/

So, the widely-used perl emblems are non-free software and maybe even
usuable for this group's T-shirts!  (Is it an official TPF Perl
Mongers group, whatever that means?)

Bad ORA!  Bad TPF!  Hurting your fans!

Annoyed by this crap,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From paulm at paulm.com  Tue Jul  1 12:29:05 2008
From: paulm at paulm.com (Paul Makepeace)
Date: Tue, 1 Jul 2008 12:29:05 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
Message-ID: <c37577dd0807010429h12a49b7cj8193189fd2435149@mail.gmail.com>

On Tue, Jul 1, 2008 at 12:20 PM, MJ Ray <mjr at phonecoop.coop> wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>> http://perl.bristolbath.org/tshirt/front.png
>> http://perl.bristolbath.org/tshirt/back.png
>
> Actually, this just came up over on http://lists.debian.org/debian-perl/
>
> That t-shirt back may be infringing the Perl Foundation's trademark.
> See http://www.perlfoundation.org/perl_trademark for their claims.
>
> The camel claims are http://www.oreillynet.com/pub/a/oreilly/perl/usage/
>
> So, the widely-used perl emblems are non-free software and maybe even
> usuable for this group's T-shirts!  (Is it an official TPF Perl
> Mongers group, whatever that means?)
>
> Bad ORA!  Bad TPF!  Hurting your fans!

It's not bad anyone. They're protecting trademarks which they are
_legally_ required to do. In practice, all you need to do is ask
permission from ORA/TPF when you have the final design and they'll
rubberstamp it. It's lightweight. No need to get het up about it.

P

>
> Annoyed by this crap,
> --
> MJ Ray (slef)
> Webmaster for hire, statistician and online shop builder for a small
> worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
> (Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>

From webmaster at cosmicperl.com  Tue Jul  1 12:59:40 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 01 Jul 2008 12:59:40 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
Message-ID: <486A1C2C.1000106@cosmicperl.com>

MJ Ray wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>   
>> http://perl.bristolbath.org/tshirt/front.png
>> http://perl.bristolbath.org/tshirt/back.png
>>     
>
> Actually, this just came up over on http://lists.debian.org/debian-perl/
>
> That t-shirt back may be infringing the Perl Foundation's trademark.
> See http://www.perlfoundation.org/perl_trademark for their claims.
>
> The camel claims are http://www.oreillynet.com/pub/a/oreilly/perl/usage/
>
> So, the widely-used perl emblems are non-free software and maybe even
> usuable for this group's T-shirts!  (Is it an official TPF Perl
> Mongers group, whatever that means?)
>
> Bad ORA!  Bad TPF!  Hurting your fans!
>
> Annoyed by this crap,
>   

When I asked O'Reilly they said they tend to give it the all clear, but 
need to see the t-shirt design first. So yes, after we all try to reach 
some kind of agreement on the t-shirts, they may have to be changed anyway.


Lyle


From webmaster at cosmicperl.com  Tue Jul  1 13:10:36 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 01 Jul 2008 13:10:36 +0100
Subject: [Bristolbathpm] Meet change
Message-ID: <486A1EBC.2040000@cosmicperl.com>

Hi All,
  As our last 2 meets clashed with Bristol Chaostreff, I suggest the 
date be changed to the 3rd Tuesday of the month rather than the last. 
This month would be missed as 3 weeks is too soon for another meet. Does 
anyone know of anything that this would clash with?


Lyle



From dave at dave.org.uk  Tue Jul  1 13:19:22 2008
From: dave at dave.org.uk (Dave Cross)
Date: Tue, 01 Jul 2008 13:19:22 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
Message-ID: <486A20CA.40502@dave.org.uk>

MJ Ray wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>> http://perl.bristolbath.org/tshirt/front.png
>> http://perl.bristolbath.org/tshirt/back.png
> 
> Actually, this just came up over on http://lists.debian.org/debian-perl/
> 
> That t-shirt back may be infringing the Perl Foundation's trademark.
> See http://www.perlfoundation.org/perl_trademark for their claims.

This was discussed on use.perl.oreg recently[1]. The bottom line is that 
TPF are just protecting their trademark - something that you have to do 
if you don't want to lose it.

TPF have never refused use of the onion logo to any PM group who have asked.

> The camel claims are http://www.oreillynet.com/pub/a/oreilly/perl/usage/

Likewise, O'Reilly are protecting their trademark. And they are, in my 
experience, very happy for PM groups to use it.

> So, the widely-used perl emblems are non-free software and maybe even
> usuable for this group's T-shirts!  (Is it an official TPF Perl
> Mongers group, whatever that means?)

Oh yes, this is an "official" group[2].

> Bad ORA!  Bad TPF!  Hurting your fans!

Not hurting anyone actually. But feel free to have a knee-jerk reaction :)

> Annoyed by this crap,

Welcome to the real world.

Cheers,

Dave...

[1] http://use.perl.org/~hex/journal/36678
[2] http://www.pm.org/groups/116.html

From adamairmailed at yahoo.com  Tue Jul  1 14:09:05 2008
From: adamairmailed at yahoo.com (adam armfield)
Date: Tue, 1 Jul 2008 06:09:05 -0700 (PDT)
Subject: [Bristolbathpm] Extending and Embedding Perl
Message-ID: <979770.18790.qm@web30307.mail.mud.yahoo.com>

Hello there,

I've got one brand new copy of Extending and Embedding Perl for sale, here are some details: 

Extending and Embedding Perl explains how to expand the functionality and usefulness of the Perl programming language and how to use Perl from C programs. It begins simply but also covers complex issues using real code examples from the Perl source. The book discusses how to write interfaces to C libraries (as well as C++ and Fortran libraries). It shows you how to implement Perl callbacks for C libraries, how to pass Perl hashes and arrays between Perl and C, and how to use the Perl Data Language infrastructure to improve the speed of array operations. 

Additionally, the book peers under the hood to see how the Perl programming language really works by looking at the interpreter. The make-up of Perl variables is discussed along with details on how a Perl program is parsed and converted to executable code. 

What's Inside: 

C For The Perl Programmer 
Basic and Advanced XS 
Embedding Perl in C Programs 
Perl Internals 
An API reference for the internal C interface to Perl 
A reference on the typemap system 
Embedding Perl into the mutt mail reader 
ABOUT THE AUTHORS...
A Perl developer for 7 years, Tim Jenness has ten modules on CPAN including two that are distributed as part of the core Perl distribution (File::Temp and Pod::LateX). 

Simon Cozens the author of Beginning Perl, was the release manager for the first four releases of the Perl 6 runtime engine, Parrot. 

Book Reviews
"My expectations were exceeded by this book in every way. This is not just a guide to writing Perl modules with C interspersed. It also provides an in depth look at everything that happens from the second Perl interprets your source file to the second it exits. ...This book fills a gap in the online Perl documentation by making many complex topics easily accessible. A good mix of theory and practice guides the reader through tasks quickly, without leaving them stranded. I recommend this book for any Perl programmers seeking to better harness Perl in their work."
?Computing Reviews, August 2003

"a welcome addition to the arsenal of Perl books. It should help with demystifying XS and associated technologies...an invaluable reference."
?Linux Journal, July 2003

"Extending and Embedding Perl provides an authoritative account of Perl's most advanced features...experienced C and Perl programmers will find this book both engaging and illuminating."
?IBM DeveloperWorks

"Extending and Embedding Perl is characterized by profound expertise, attention to detail, literate writing, and excellent editorial and production values. This book is so necessary that it would be a must-have for the problem domain, even were it less finely crafted?a masterpiece that will keep its value for years...an exceptionally useful and readable book."
?The Perl Journal, November 2002


"?a deliberate thorough guide led by authors that are both extremely knowledgeable and also capable of communicating that knowledge...?
?Kansas City Perl Mongers


"...aims to take the black magic out of understanding [Perl]...indispensable in gaining a good foothold on using Perl in, from, and around C."
?Slashdot.org


"a must-have for the problem domain, even were it less finely crafted; we're lucky that the one book to adequately address the most puzzling facets of Perl happens to be a masterpiece that will keep its value for years."
?The Perl Journal

"I've not seen a more complete and well written source of information on the subject of embedding...Each chapter has an extensive set of annotated examples and the style of writing makes the learning process easier...The book is great for learning and great for reference, which is a combination other books often fail to achieve."
?Perlmonks

"Everything related to Perl internals is found here; the Perl internal API, XS, SWIG, embedding Perl in C, and C in Perl. The book is organized well, and is fine as a reference work."
?The Perl Review

"This book was indispensable in gaining a good foot-hold on using Perl in, from, and around C. I found it a good reference guide as well as an easy linear read. It is not a replacement for the perlguts, perlapi and perlxs documentation. But then again, it doesn't try to be. The annotated code examples with every line explained make following the book with development of your own solution a lot easier than in some other books."
?Use.Perl.org

"This book is well written, concise, and an absolute must for all those, such as myself, who need to work with C and Perl, but who've been putting off the ... day. With this superb book, we cowards no longer have any excuses. Magic! :-)"
?Andrew Duncan, Amazon Reader

Paperback with 384 pages.

As I say, this book is brand new, full UK retail price is around ?40, Amazon sell it for ?25, I'm selling this copy for just ?15.

The buyer can collect from me in central Bristol, or I can post this for an additional charge.

Please email off list if you'd like to buy this.

All the best

Adam 


      __________________________________________________________
Not happy with your email address?.
Get the one you really want - millions of new email addresses available now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html


From webmaster at cosmicperl.com  Tue Jul  1 14:18:06 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 01 Jul 2008 14:18:06 +0100
Subject: [Bristolbathpm] Extending and Embedding Perl
In-Reply-To: <979770.18790.qm@web30307.mail.mud.yahoo.com>
References: <979770.18790.qm@web30307.mail.mud.yahoo.com>
Message-ID: <486A2E8E.1050107@cosmicperl.com>

I'll have it!

CosmicPerl.com CGI Scripts <http://www.cosmicperl.com> 	Lyle Hopkins
"How can I help?"



adam armfield wrote:
> Hello there,
>
> I've got one brand new copy of Extending and Embedding Perl for sale, here are some details: 
>
> Extending and Embedding Perl explains how to expand the functionality and usefulness of the Perl programming language and how to use Perl from C programs. It begins simply but also covers complex issues using real code examples from the Perl source. The book discusses how to write interfaces to C libraries (as well as C++ and Fortran libraries). It shows you how to implement Perl callbacks for C libraries, how to pass Perl hashes and arrays between Perl and C, and how to use the Perl Data Language infrastructure to improve the speed of array operations. 
>
> Additionally, the book peers under the hood to see how the Perl programming language really works by looking at the interpreter. The make-up of Perl variables is discussed along with details on how a Perl program is parsed and converted to executable code. 
>
> What's Inside: 
>
> C For The Perl Programmer 
> Basic and Advanced XS 
> Embedding Perl in C Programs 
> Perl Internals 
> An API reference for the internal C interface to Perl 
> A reference on the typemap system 
> Embedding Perl into the mutt mail reader 
> ABOUT THE AUTHORS...
> A Perl developer for 7 years, Tim Jenness has ten modules on CPAN including two that are distributed as part of the core Perl distribution (File::Temp and Pod::LateX). 
>
> Simon Cozens the author of Beginning Perl, was the release manager for the first four releases of the Perl 6 runtime engine, Parrot. 
>
> Book Reviews
> "My expectations were exceeded by this book in every way. This is not just a guide to writing Perl modules with C interspersed. It also provides an in depth look at everything that happens from the second Perl interprets your source file to the second it exits. ...This book fills a gap in the online Perl documentation by making many complex topics easily accessible. A good mix of theory and practice guides the reader through tasks quickly, without leaving them stranded. I recommend this book for any Perl programmers seeking to better harness Perl in their work."
> ?Computing Reviews, August 2003
>
> "a welcome addition to the arsenal of Perl books. It should help with demystifying XS and associated technologies...an invaluable reference."
> ?Linux Journal, July 2003
>
> "Extending and Embedding Perl provides an authoritative account of Perl's most advanced features...experienced C and Perl programmers will find this book both engaging and illuminating."
> ?IBM DeveloperWorks
>
> "Extending and Embedding Perl is characterized by profound expertise, attention to detail, literate writing, and excellent editorial and production values. This book is so necessary that it would be a must-have for the problem domain, even were it less finely crafted?a masterpiece that will keep its value for years...an exceptionally useful and readable book."
> ?The Perl Journal, November 2002
>
>
> "?a deliberate thorough guide led by authors that are both extremely knowledgeable and also capable of communicating that knowledge...?
> ?Kansas City Perl Mongers
>
>
> "...aims to take the black magic out of understanding [Perl]...indispensable in gaining a good foothold on using Perl in, from, and around C."
> ?Slashdot.org
>
>
> "a must-have for the problem domain, even were it less finely crafted; we're lucky that the one book to adequately address the most puzzling facets of Perl happens to be a masterpiece that will keep its value for years."
> ?The Perl Journal
>
> "I've not seen a more complete and well written source of information on the subject of embedding...Each chapter has an extensive set of annotated examples and the style of writing makes the learning process easier...The book is great for learning and great for reference, which is a combination other books often fail to achieve."
> ?Perlmonks
>
> "Everything related to Perl internals is found here; the Perl internal API, XS, SWIG, embedding Perl in C, and C in Perl. The book is organized well, and is fine as a reference work."
> ?The Perl Review
>
> "This book was indispensable in gaining a good foot-hold on using Perl in, from, and around C. I found it a good reference guide as well as an easy linear read. It is not a replacement for the perlguts, perlapi and perlxs documentation. But then again, it doesn't try to be. The annotated code examples with every line explained make following the book with development of your own solution a lot easier than in some other books."
> ?Use.Perl.org
>
> "This book is well written, concise, and an absolute must for all those, such as myself, who need to work with C and Perl, but who've been putting off the ... day. With this superb book, we cowards no longer have any excuses. Magic! :-)"
> ?Andrew Duncan, Amazon Reader
>
> Paperback with 384 pages.
>
> As I say, this book is brand new, full UK retail price is around ?40, Amazon sell it for ?25, I'm selling this copy for just ?15.
>
> The buyer can collect from me in central Bristol, or I can post this for an additional charge.
>
> Please email off list if you'd like to buy this.
>
> All the best
>
> Adam 
>
>
>       __________________________________________________________
> Not happy with your email address?.
> Get the one you really want - millions of new email addresses available now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>
>   

From nigel at turbo10.com  Tue Jul  1 14:38:39 2008
From: nigel at turbo10.com (Nigel Hamilton)
Date: Tue, 1 Jul 2008 14:38:39 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486A20CA.40502@dave.org.uk>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
	<486A20CA.40502@dave.org.uk>
Message-ID: <50fec4060807010638u5a50d2a9n1d0df60475332982@mail.gmail.com>

Hi Lyle,

    I think it is easier to drop "the Camel" and just use the TPF "Onion" as
it means the goodwill accrues only to the custodian of Perl's IP and not a
third party [1].

    Ultimately I still think Perl needs a 'lovemark' but I'm really happy to
see the interlocking branding of Perl6 - with sub-brands (rakudo, pugs etc).
It's great to see that since I made this proposal[2] the TPF are starting to
improve trade mark protection for Perl.


Nige

[1] http://blog.trexy.com/2006/07/lovemarkr-for-perl.html
[2] YAPC::NA 2006 - "Lovemark" Proposal to the TPF :
http://thegoo.org/love-marks-for-perl.pdf
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080701/6d8dff25/attachment.html 

From mjr at phonecoop.coop  Tue Jul  1 14:52:33 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Tue, 01 Jul 2008 14:52:33 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486A20CA.40502@dave.org.uk>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
	<486A20CA.40502@dave.org.uk>
Message-ID: <486a36a1.++RqGcmpTlTNjoB8%mjr@phonecoop.coop>

Dave Cross <dave at dave.org.uk> wrote:
> MJ Ray wrote: [...]
> > Actually, this just came up over on http://lists.debian.org/debian-perl/
> > 
> > That t-shirt back may be infringing the Perl Foundation's trademark.
> > See http://www.perlfoundation.org/perl_trademark for their claims.
>
> This was discussed on use.perl.oreg recently[1]. The bottom line is that 
> TPF are just protecting their trademark - something that you have to do 
> if you don't want to lose it.
> [1] http://use.perl.org/~hex/journal/36678

That's often claimed, but what evidence is there that holders have to
license their trademarks in ways incompatible with free software and
free culture uses of them?

Note the near-total lack of references given in that use.perl thread.
Everyone thinks they know trademark law, but it looks like the same
old legends repeated.

Those comments suggest that Allison Randal drafted the trademark
licence.  Following links to
http://www.onyxneon.com/people/allison.html suggests she's not a
lawyer.  Was a lawyer consulted and what were they asked? (IIRC, we're
probably not going to be allowed to know the lawyer's answer, but we
can be told the question.)

Sun play nicer with OpenJDK and the Duke logo (IIRC), so why not Perl?

Meanwhile, lisp has the round brackets and scheme has lambda and 
neither community seems to feel the need to trademark-enforce them.
Is Perl more endangered than Lisp and Scheme now?

[...]
> > Bad ORA!  Bad TPF!  Hurting your fans!
>
> Not hurting anyone actually. But feel free to have a knee-jerk reaction :)

Sorry, I forgot that not everyone on this list sees being able to do
things without begging "intellectual property" owners for permission
as an obviously good thing.  That's a debate for another time/place.

> > Annoyed by this crap,
>
> Welcome to the real world.

No, restrictive trademarks aren't real.  This is pain invented by
overlegislation and protectionism.

Regards,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From david at cantrell.org.uk  Tue Jul  1 15:14:43 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Tue, 1 Jul 2008 15:14:43 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
	things from a different angle...
In-Reply-To: <16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>
References: <48695161.1030201@cosmicperl.com>
	<16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>
Message-ID: <20080701141443.GK19214@bytemark.barnyard.co.uk>

On Tue, Jul 01, 2008 at 09:13:31AM +0100, Alex Francis wrote:

> if you find yourself having to look for an experienced Perl
> programmer, well, good luck.

My lovely employer was recently looking for an experienced perl person
who could also do sysadminning and touchy-feely customer-facing stuff
too.  We had quite a few good CVs, from a combination of pimps,
http://jobs.perl.org, and jobs at london.pm.org.

It's not that hard.  You do have to be willing to pay enough though, and
accept that you'll never find someone who *exactly* matches your very
long wishlist of skills and experience.

That said, I suspect that it would be a lot harder to find such people
who are willing to live and work outside London.

-- 
David Cantrell | Enforcer, South London Linguistic Massive

  Languages for which ISO-Latin-$n is not necessary, #1 in a series:

    Latin

From revford at blueyonder.co.uk  Tue Jul  1 16:22:59 2008
From: revford at blueyonder.co.uk (gav)
Date: Tue, 1 Jul 2008 16:22:59 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <486a36a1.++RqGcmpTlTNjoB8%mjr@phonecoop.coop>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
	<486A20CA.40502@dave.org.uk>
	<486a36a1.++RqGcmpTlTNjoB8%mjr@phonecoop.coop>
Message-ID: <20080701152258.GA1894@bluemidget.local>

On Tue, Jul 01, 2008 at 02:52:33PM +0100, MJ Ray wrote:
> No, restrictive trademarks aren't real.  This is pain invented by
> overlegislation and protectionism.


Maybe I'm missing something here, but wouldn't it be as simple as TPF using a
Creative Commons licence for the logo?

http://creativecommons.org/license/

Something like Attribution, Non-Commercial and Share Alike?

-- 
Gav Ford
revford at blueyonder.co.uk 
http://revford.pwp.blueyonder.co.uk 
I think we need to:  Attach the graviton deflector
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080701/86b696a7/attachment.bin 

From mjr at phonecoop.coop  Tue Jul  1 17:11:25 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Tue, 01 Jul 2008 17:11:25 +0100
Subject: [Bristolbathpm] T-Shirt... and you thought I'd forgotten ;)
In-Reply-To: <20080701152258.GA1894@bluemidget.local>
References: <4868F684.8030901@cosmicperl.com>
	<486a1317.O53WCB5APvcTvEfy%mjr@phonecoop.coop>
	<486A20CA.40502@dave.org.uk>
	<486a36a1.++RqGcmpTlTNjoB8%mjr@phonecoop.coop>
	<20080701152258.GA1894@bluemidget.local>
Message-ID: <486a572d.rYvSyZ2dOJEfmR+E%mjr@phonecoop.coop>

"gav" <revford at blueyonder.co.uk> wrote:
> On Tue, Jul 01, 2008 at 02:52:33PM +0100, MJ Ray wrote:
> > No, restrictive trademarks aren't real.  This is pain invented by
> > overlegislation and protectionism.
>
> Maybe I'm missing something here, but wouldn't it be as simple as TPF using a
> Creative Commons licence for the logo? [...]
> Something like Attribution, Non-Commercial and Share Alike?

I think most of the Creative Commons licences are copyright licences,
with a few patent ones in there (ccGPL3?).  It's the onion and camel
trademarks which are problems and wouldn't be covered by a copyright
licence. I'd not even considered whether copyright is a problem yet!

Non-Commercial would make t-shirt sales impossible without obtaining
more permission.

Hope that explains,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From webmaster at cosmicperl.com  Tue Jul  1 18:34:40 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 01 Jul 2008 18:34:40 +0100
Subject: [Bristolbathpm] T-Shirt cont...
Message-ID: <486A6AB0.9020200@cosmicperl.com>

Hi All,
  Starting a new thread as the old ones gone on a bit of a trademark 
tangent.

Judging by the responses so far, I don't think everyone will be totally 
happy with just one slogan. Thinking back to my brothers stag the 
t-shirts I got for everyone were able to have individual slogans (and 
names, but I'm not sure how'd it look if we all had our own names on the 
shirts).

So how about we all have B&BPM logo, and URL on the front, and choose 
our own slogans for the back :)


Lyle



From paulm at paulm.com  Tue Jul  1 19:47:10 2008
From: paulm at paulm.com (Paul Makepeace)
Date: Tue, 1 Jul 2008 19:47:10 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
	things from a different angle...
In-Reply-To: <48695161.1030201@cosmicperl.com>
References: <48695161.1030201@cosmicperl.com>
Message-ID: <c37577dd0807011147i322dc790t2df039fe9158fee5@mail.gmail.com>

There's a lot to be said for standardising within an organisation.
That said, realistically there's stuff that C++, Java, JavaScript, and
C (to name some 'big' languages) do that Perl just can't do. Of
course, if you can say your business will never need to write software
that runs extremely fast, be bounded in memory use, link to a kernel,
build UIs, run in a web browser, or do well, whatever Java does :-)
then fair enough.

On a tangent,
The old idea that a decent programmer can pick up a new language and
be solidly productive in it in a number of weeks is a hoary old load
of bollocks if you ask me. Sure they can produce _code_ but it won't
be very good and it'll take them much longer than someone skilled in
it, and no programmer I've ever met can start producing complex
application-class code because it simply takes time to learn and
become familiar with libraries (often the largest task), and the
quirks of debugging, i.e. what errors look and feel like in different
situations and how that maps to the actual problem.

I'd consider myself a decent programmer, and python a reasonably
learnable language - I've been coding in it on & off for two years and
I'm still learning stuff, and I'm not talking about bizarre esoteric
edge cases. I've been coding perl pretty much continuously since '94
and am still learning ideas & libraries that make me more useful, and
that's true of pretty much everyone I've spoken to. You just can't get
away from it taking gobs of time & effort to get good at this stuff.

(This is all in support of your basic premise there's real value is
specialising, btw.)

Paul

On Mon, Jun 30, 2008 at 10:34 PM, Lyle <webmaster at cosmicperl.com> wrote:
>
> Hi All,
>  These days I hear a lot of:-
> "You should use language X for this, language Y for that and language Z
> for the other"
> "Because language X was designed for this, Y is most commonly used for
> that, and Z was made for that OS"
>
> Sounds perfectly fine in theory, but what is the reality?
>
> The way I see it this method has many problems for businesses,
> particularly SME's but I see big businesses being effected as well.
>
> Here is a hypothetical Scenario.
>
> You hire 'Super Programmer' who can code X Y Z at a high level. He makes
> you sysadmin scripts in X, web scripts in Y and gets other pieces of
> software talking to each other with Z. These guys are rare, making them
> difficult to replace by a single person when they leave. They are also
> very expensive.
>
> He leaves the company and you cannot find a replacement who knows all of
> the languages at a high level, you need someone fast and it'd take to
> long to train someone up.
> You end up employing 2 people instead one doing X and Y and another
> doing Z. This means more managerial time is devoted to getting the two
> programmers working together properly and more admin time processing
> things like PAYE, etc.
>
> A month on and one of your programmers finds that some of the software
> he needs to update and maintain is actually coded in rare language W.
> The original 'super programmer' discovered this new language and used it
> as it was designed specifically for the job in hand. Unfortunately there
> is little documentation and the language has since died and is no longer
> maintained. Much time is lost recoding everything to language Z.
>
> The companies business shifts a little and they want to target a new
> platform. The best languages to use on this new platform are A, B and C.
> The current programmers are not familiar with these languages. Training
> them would be to slow and costly. It's feared that them having to use to
> many different languages, will hinder their performance on the ones they
> are currently using.
>
> So the decision is made to outsource this platform to another company.
> Many issues arise getting the outsourced programmers to work well with
> the in house programmers, and keeping both platform versions in sync
> with each other.
>
> A lot of time. A lot of money. A lot of headache.
>
>
> Here is another hypothetical Scenario.
>
> You hire a Perl programmer. He makes you sysadmin scripts in Perl, web
> scripts in Perl and gets other pieces of software talking to each other
> with Perl.
>
> When he leaves you replace him with another Perl programmer.
>
> When the work gets to much for him, you hire another Perl programmer to
> work with him.
>
> When the company wants to target another platform, your Perl programmers
> code the software in Perl.
>
> Of course coding in Perl all the programmers get to go home early :)
>
> Call me a Perl loving fool if you will. But I don't see any other
> scripting language out there that comes close to handling a diverse
> range of tasks, as well as Perl does.
>
>
> I apologise that the scenario isn't presented better. My English was
> never that good :p
>
>
>
> Lyle
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm

From webmaster at cosmicperl.com  Tue Jul  1 22:17:03 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 01 Jul 2008 22:17:03 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
 things from a different angle...
In-Reply-To: <16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>
References: <48695161.1030201@cosmicperl.com>
	<16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>
Message-ID: <486A9ECF.6080700@cosmicperl.com>

Alex Francis wrote:
> On Mon, Jun 30, 2008 at 10:34 PM, Lyle <webmaster at cosmicperl.com> wrote:
>   
>> Hi All,
>>  These days I hear a lot of:-
>> "You should use language X for this, language Y for that and language Z
>> for the other"
>> "Because language X was designed for this, Y is most commonly used for
>> that, and Z was made for that OS"
>>
>>     
>
> Hi Lyle
>
> While I'm (of course) very much in favour of Perl, I think for the
> most part it's possible to hire bright programmers who can adapt to
> whatever language or platform they need to. I've found the bigger
> challenge to be in helping non-technical senior management to make
> sensible choices and in helping them to trust their expensive
> technical staff.
>   

I think your right that you can get bright programmers to adapt. But I 
think it'll take much dedication to be able to program at a high level 
in any new language.

> I've worked places where the language/platform selection has been
> based on some made-up nonsense about customers being more comfortable
> with X or Y and if we use Z then we won't be able to sell (Z being
> anything that isn't Java or .net). I've also seen what should have
> been considered strategic decisions become panic-stricken knee-jerk
> changes which waste huge amounts of money.
>   

There does seem a big business thing to automatically assume .NET or 
Java are the best solution. I think a lot of this comes from SUNs and MS 
massive certification systems. But Perl certification is a whole other 
argument we'll leave for another day :)

> Something I would say about Perl is that as a programmer it exposes
> you to a lot of concepts that make other languages easier to pick up
> (when you inevitably have to pick them up).

I'd agree. When I went back to look at some C++ stuff, all the Perl I've 
done made things a lot clearer. Although C# looks totally weird to me, 
and I dread it when a customer has a c# shopping cart. Guess at some 
point I'm going to have to learn some Java.... But I just don't want to :/



Lyle


From mjr at phonecoop.coop  Wed Jul  2 09:59:08 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Wed, 02 Jul 2008 09:59:08 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
 things from a different angle...
In-Reply-To: <486A9ECF.6080700@cosmicperl.com>
References: <48695161.1030201@cosmicperl.com>
	<16afc9550807010113o10442a95v45237caeaa90de20@mail.gmail.com>
	<486A9ECF.6080700@cosmicperl.com>
Message-ID: <486b435c.7mN21xpV9OKs5yJQ%mjr@phonecoop.coop>

Three replies in one - hope that's OK.

David Cantrell <david at cantrell.org.uk> wrote:
> That said, I suspect that it would be a lot harder to find such people
> who are willing to live and work outside London.

That surprises me.  There seem to be lots of good people living
outside London who really don't want to work inside it with the
overcrowding and the black snot and so on.  I think there could a
London/non-London split - I know people both inside and outside London
and I don't expect many of them to be moving across the boundary any
time soon.

"Paul Makepeace" <paulm at paulm.com> wrote:
> The old idea that a decent programmer can pick up a new language and
> be solidly productive in it in a number of weeks is a hoary old load
> of bollocks if you ask me. Sure they can produce _code_ but it won't
> be very good and it'll take them much longer than someone skilled in
> it, and no programmer I've ever met can start producing complex
> application-class code because [...]

I've met some who can start producing complex application-class code
fairly quickly, but not starting from a blank page.  90%+ of our work
is developing existing applications, so that's not a huge problem.

Even so, I feel it's easy to overlook the need to assign at least one
more experienced programmer to help them get started, through
something like pair programming, or at least close reviews and
feedback.  Maybe companies who don't have worker-learning as part of
their Mission find it harder to justify that level investment in their
workers, so prefer to hire ready-made programmers?

Lyle <webmaster at cosmicperl.com> wrote: [...]
> There does seem a big business thing to automatically assume .NET or 
> Java are the best solution. I think a lot of this comes from SUNs and MS 
> massive certification systems. But Perl certification is a whole other 
> argument we'll leave for another day :)

Not just Perl.  Certification entirely is a whole other argument.  I
don't trust many other organisations enough to delegate checking
programmer competence to them.  It surprises me so many do and doesn't
surprise me at all that so many get into pickles.  I guess "no-one got
fired for believing EdExcel/Sun/University of Bottles" is a new
variation on an old meme.

Regards,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From webmaster at cosmicperl.com  Wed Jul  2 21:16:33 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 02 Jul 2008 21:16:33 +0100
Subject: [Bristolbathpm] Which language to use for what? - Looking at
 things from a different angle...
In-Reply-To: <c37577dd0807011147i322dc790t2df039fe9158fee5@mail.gmail.com>
References: <48695161.1030201@cosmicperl.com>
	<c37577dd0807011147i322dc790t2df039fe9158fee5@mail.gmail.com>
Message-ID: <486BE221.5090103@cosmicperl.com>

Paul Makepeace wrote:
> There's a lot to be said for standardising within an organisation.
> That said, realistically there's stuff that C++, Java, JavaScript, and
> C (to name some 'big' languages) do that Perl just can't do. Of
> course, if you can say your business will never need to write software
> that runs extremely fast, be bounded in memory use, link to a kernel,
> build UIs, run in a web browser, or do well, whatever Java does :-)
> then fair enough.
>   

That's exactly why I updated my message before sending to be scripting 
language rather than just language.

> But I don't see any other
> scripting language out there that comes close to handling a diverse
> range of tasks, as well as Perl does.


I accept there are things you just have to do in C, C++ and JavaScript. 
Or I even guess Java for things like phone games (though I've never 
really liked Java for some reason).


> (This is all in support of your basic premise there's real value is
> specialising, btw.)
>   

Yes. Real value in specialising in Perl!



Lyle


From webmaster at cosmicperl.com  Fri Jul  4 01:25:57 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 04 Jul 2008 01:25:57 +0100
Subject: [Bristolbathpm] T-Shirt cont...
In-Reply-To: <486A6AB0.9020200@cosmicperl.com>
References: <486A6AB0.9020200@cosmicperl.com>
Message-ID: <486D6E15.30801@cosmicperl.com>

Lyle wrote:
> Hi All,
>   Starting a new thread as the old ones gone on a bit of a trademark 
> tangent.
>
> Judging by the responses so far, I don't think everyone will be totally 
> happy with just one slogan. Thinking back to my brothers stag the 
> t-shirts I got for everyone were able to have individual slogans (and 
> names, but I'm not sure how'd it look if we all had our own names on the 
> shirts).
>
> So how about we all have B&BPM logo, and URL on the front, and choose 
> our own slogans for the back :

Shameless bump... :)


Lyle


From pmh at edison.ioppublishing.com  Fri Jul  4 10:08:42 2008
From: pmh at edison.ioppublishing.com (Peter Haworth)
Date: Fri,  4 Jul 2008 10:08:42 +0100
Subject: [Bristolbathpm] T-Shirt cont...
References: <486A6AB0.9020200@cosmicperl.com> <486D6E15.30801@cosmicperl.com>
Message-ID: <PGM.20080704090842.7623.9102@edison.ioppublishing.com>

An embedded and charset-unspecified text was scrubbed...
Name: not available
Url: http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080704/439c2af3/attachment.pl 

From webmaster at cosmicperl.com  Fri Jul  4 17:45:13 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 04 Jul 2008 17:45:13 +0100
Subject: [Bristolbathpm] T-Shirt cont...
In-Reply-To: <PGM.20080704090842.7623.9102@edison.ioppublishing.com>
References: <486A6AB0.9020200@cosmicperl.com> <486D6E15.30801@cosmicperl.com>
	<PGM.20080704090842.7623.9102@edison.ioppublishing.com>
Message-ID: <486E5399.2020300@cosmicperl.com>

Peter Haworth wrote:
> Actually, shouldn't we be able to use bristolbath.pm.org?
>   

Well, we've got bristol.pm and bath.pm, but not bristolbath.pm

So potentially the people from bristol could have bristol.pm...
and the people from bath could have bath.pm...
Or you could just have perl.bristolbath.org

I guess I could apply to see if they would create bristolbath.pm for us. 
I think the problem before was that mailman doesn't let you rename a 
list, as I was given bath and bristol pm it worked out easier this way.


Lyle


From webmaster at cosmicperl.com  Mon Jul  7 11:58:24 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Mon, 07 Jul 2008 11:58:24 +0100
Subject: [Bristolbathpm] Meet change
In-Reply-To: <486A1EBC.2040000@cosmicperl.com>
References: <486A1EBC.2040000@cosmicperl.com>
Message-ID: <4871F6D0.408@cosmicperl.com>

Lyle wrote:
> Hi All,
>   As our last 2 meets clashed with Bristol Chaostreff, I suggest the 
> date be changed to the 3rd Tuesday of the month rather than the last. 
> This month would be missed as 3 weeks is too soon for another meet. Does 
> anyone know of anything that this would clash with?
>
>
> Lyle
>   

I guess not. Next meet 19th August, Bristol ;)


Lyle


From mjr at phonecoop.coop  Mon Jul  7 12:58:48 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Mon, 07 Jul 2008 12:58:48 +0100
Subject: [Bristolbathpm] T-Shirt cont...
In-Reply-To: <486D6E15.30801@cosmicperl.com>
References: <486A6AB0.9020200@cosmicperl.com> <486D6E15.30801@cosmicperl.com>
Message-ID: <487204f8.WfsvmJ72noXSW9XS%mjr@phonecoop.coop>

Lyle <webmaster at cosmicperl.com> wrote:
> Lyle wrote:
> > Hi All,
> >   Starting a new thread as the old ones gone on a bit of a trademark 
> > tangent.
> >
> > Judging by the responses so far, I don't think everyone will be totally 
> > happy with just one slogan. Thinking back to my brothers stag the 
> > t-shirts I got for everyone were able to have individual slogans (and 
> > names, but I'm not sure how'd it look if we all had our own names on the 
> > shirts).
> >
> > So how about we all have B&BPM logo, and URL on the front, and choose 
> > our own slogans for the back :
>
> Shameless bump... :)

Erm, well, how about it?

I'm not sure what's wanted in reply.

Regards,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From webmaster at cosmicperl.com  Mon Jul  7 13:38:06 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Mon, 07 Jul 2008 13:38:06 +0100
Subject: [Bristolbathpm] T-Shirt cont...
In-Reply-To: <487204f8.WfsvmJ72noXSW9XS%mjr@phonecoop.coop>
References: <486A6AB0.9020200@cosmicperl.com> <486D6E15.30801@cosmicperl.com>
	<487204f8.WfsvmJ72noXSW9XS%mjr@phonecoop.coop>
Message-ID: <48720E2E.10208@cosmicperl.com>

MJ Ray wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>   
>> Lyle wrote:
>>     
>>> So how about we all have B&BPM logo, and URL on the front, and choose 
>>> our own slogans for the back :
>>>       
>> Shameless bump... :)
>>     
>
> Erm, well, how about it?
>
> I'm not sure what's wanted in reply.
>   

Yes I'd go with that or no I don't like the idea... Pretty much like 
Peters response.


Lyle


From nigel at turbo10.com  Mon Jul  7 13:45:27 2008
From: nigel at turbo10.com (Nigel Hamilton)
Date: Mon, 7 Jul 2008 13:45:27 +0100
Subject: [Bristolbathpm] Meet change
In-Reply-To: <4871F6D0.408@cosmicperl.com>
References: <486A1EBC.2040000@cosmicperl.com> <4871F6D0.408@cosmicperl.com>
Message-ID: <50fec4060807070545u43d62079ud4c55812d35536bc@mail.gmail.com>

> I guess not. Next meet 19th August, Bristol ;)

OK. Sounds good. Looking forward to it.

Nige
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080707/adfd912d/attachment.html 

From webmaster at cosmicperl.com  Tue Jul  8 00:22:18 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 08 Jul 2008 00:22:18 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
Message-ID: <4872A52A.4030503@cosmicperl.com>

Hi All,
 I've hit a bit of a snag with restoring from a mysql backup. Having 
posted to a MySQL forum with no reply, I'm thinking the solution will be 
to use a bit of perl to get things working properly.

I wont go into too much detail, but the basic scenario is this:-

Database A is the old one full of data
Database B is the new one with all the same tables, starting to be 
filled with new data.

A mysqldump of database A has been made with the command
mysqldump --opt --skip-add-drop-table --no-create-info --complete-insert 
-uuser -ppass -hlocalhost databasename > ./sampledump.bak

So there are no drop tables, etc.

Some of the tables act as tallies for pieces of data, such:-
table ( ID (unique), totalamount )

How can I restore from the backup so that these table rows are merged 
rather than replaced?

For example:-
The backup has:-
user1 (ID), 1000 (totalamount)

new database has
user1 (ID), 100 (totalamount)

I want to restore the backup and get:-
user1, 1100


I still have the old database so potentially I can make a new mysqldump 
if needed.


The important bit of code in the mysqldump is:-

LOCK TABLES `tally` WRITE;
/*!40000 ALTER TABLE `tally` DISABLE KEYS */;
INSERT INTO `tally` (`ID`, `totalamount`) VALUES 
('user1','1000'),('user2','2000'),('user3','3000');
/*!40000 ALTER TABLE `tally` ENABLE KEYS */;
UNLOCK TABLES;


My SQL isn't that great so there may be a simple SQL trick I'm missing...

At the moment I'm thinking it would be too much of a pain to try and 
extract from this file which ones needs to be updated and which just 
inserted. I'm thinking maybe re-create all the tables with a prefix, add 
a prefix to the tables in this dump so I can get all the data into the 
new database without disturbing the current database. Then move the non 
tally data from one table to another, and finally tally up the data that 
needs to be tallied. Seems a bit long winded...

Hoping someone here may have experience with this kind of thing... Or a 
better idea...



Lyle



From nigel at turbo10.com  Tue Jul  8 08:16:09 2008
From: nigel at turbo10.com (Nigel Hamilton)
Date: Tue, 8 Jul 2008 08:16:09 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <4872A52A.4030503@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com>
Message-ID: <50fec4060807080016p73d1389bh7bbc3e7a7af08866@mail.gmail.com>

HI Lyle,

    Your last idea sounds like a good one. Add the new tables with a suffix
.new and then rename them when you're happy. Once all the tables are in the
same database you can run UPDATE commands to increase the total columns - or
whatever cronjob is computing the totals in the first place might magically
recalculate them for you? ;-)

    I use webmin's MySQL module for doing this sort of thing as it helps you
with RENAME and other less common SQL you otherwise have to get the manual
for.

Nige



2008/7/8 Lyle <webmaster at cosmicperl.com>:

> Hi All,
>  I've hit a bit of a snag with restoring from a mysql backup. Having
> posted to a MySQL forum with no reply, I'm thinking the solution will be
> to use a bit of perl to get things working properly.
>
> I wont go into too much detail, but the basic scenario is this:-
>
> Database A is the old one full of data
> Database B is the new one with all the same tables, starting to be
> filled with new data.
>
> A mysqldump of database A has been made with the command
> mysqldump --opt --skip-add-drop-table --no-create-info --complete-insert
> -uuser -ppass -hlocalhost databasename > ./sampledump.bak
>
> So there are no drop tables, etc.
>
> Some of the tables act as tallies for pieces of data, such:-
> table ( ID (unique), totalamount )
>
> How can I restore from the backup so that these table rows are merged
> rather than replaced?
>
> For example:-
> The backup has:-
> user1 (ID), 1000 (totalamount)
>
> new database has
> user1 (ID), 100 (totalamount)
>
> I want to restore the backup and get:-
> user1, 1100
>
>
> I still have the old database so potentially I can make a new mysqldump
> if needed.
>
>
> The important bit of code in the mysqldump is:-
>
> LOCK TABLES `tally` WRITE;
> /*!40000 ALTER TABLE `tally` DISABLE KEYS */;
> INSERT INTO `tally` (`ID`, `totalamount`) VALUES
> ('user1','1000'),('user2','2000'),('user3','3000');
> /*!40000 ALTER TABLE `tally` ENABLE KEYS */;
> UNLOCK TABLES;
>
>
> My SQL isn't that great so there may be a simple SQL trick I'm missing...
>
> At the moment I'm thinking it would be too much of a pain to try and
> extract from this file which ones needs to be updated and which just
> inserted. I'm thinking maybe re-create all the tables with a prefix, add
> a prefix to the tables in this dump so I can get all the data into the
> new database without disturbing the current database. Then move the non
> tally data from one table to another, and finally tally up the data that
> needs to be tallied. Seems a bit long winded...
>
> Hoping someone here may have experience with this kind of thing... Or a
> better idea...
>
>
>
> Lyle
>
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080708/16628101/attachment.html 

From psykx.out at googlemail.com  Wed Jul  9 08:06:49 2008
From: psykx.out at googlemail.com (psykx.out)
Date: Wed, 09 Jul 2008 08:06:49 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <4872A52A.4030503@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com>
Message-ID: <48746389.2070704@googlemail.com>

I'm not I understand perfectly, you can use 'replace into' instead of 
'insert into' if you want to overwrite (edit) existing data. replace 
acts like insert if the is no existing record.

Max

From mhsparks at gmail.com  Wed Jul  9 12:47:52 2008
From: mhsparks at gmail.com (Mark Hughes)
Date: Wed, 9 Jul 2008 12:47:52 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <4872A52A.4030503@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com>
Message-ID: <ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>

On Tue, Jul 8, 2008 at 12:22 AM, Lyle <webmaster at cosmicperl.com> wrote:
> How can I restore from the backup so that these table rows are merged
> rather than replaced?
>
> For example:-
> The backup has:-
> user1 (ID), 1000 (totalamount)
>
> new database has
> user1 (ID), 100 (totalamount)
>
> I want to restore the backup and get:-
> user1, 1100
>

Hi Lyle,

You could try the 'INSERT ... ON DUPLICATE KEY UPDATE' syntax

http://dev.mysql.com/doc/refman/5.0/en/insert-on-duplicate.html

Something like:

insert into new_table  select * from old_table on duplicate key update
new_table.totalamount = new_table.totalamount + old_table.totalamount;

Or if old_table doesn't have any unique rows, or you don't want to add
them to new_table just use a multiple table update

update new_table  set new_table.totalamount = new_table.totalamount +
old_table.totalamount where new_table.id = old_table.id;

Whatever you try, have a backup and do it locally first :-)

Mark.

From webmaster at cosmicperl.com  Wed Jul  9 22:46:36 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 09 Jul 2008 22:46:36 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <50fec4060807080016p73d1389bh7bbc3e7a7af08866@mail.gmail.com>
References: <4872A52A.4030503@cosmicperl.com>
	<50fec4060807080016p73d1389bh7bbc3e7a7af08866@mail.gmail.com>
Message-ID: <487531BC.90600@cosmicperl.com>

Nigel Hamilton wrote:
> HI Lyle,
>
>     Your last idea sounds like a good one. Add the new tables with a 
> suffix .new and then rename them when you're happy. Once all the 
> tables are in the same database you can run UPDATE commands to 
> increase the total columns - or whatever cronjob is computing the 
> totals in the first place might magically recalculate them for you? ;-)

I think your right. Once the backup data is in the new database I'll be 
able to sort it all out.

>  
>     I use webmin's MySQL module for doing this sort of thing as it 
> helps you with RENAME and other less common SQL you otherwise have to 
> get the manual for.

I've always found Webmins MySQL module a bit of a pain and usually end 
up installing phpMyAdmin.


Lyle

From webmaster at cosmicperl.com  Wed Jul  9 22:47:29 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 09 Jul 2008 22:47:29 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <48746389.2070704@googlemail.com>
References: <4872A52A.4030503@cosmicperl.com> <48746389.2070704@googlemail.com>
Message-ID: <487531F1.6040901@cosmicperl.com>

psykx.out wrote:
> I'm not I understand perfectly, you can use 'replace into' instead of 
> 'insert into' if you want to overwrite (edit) existing data. replace 
> acts like insert if the is no existing record

Thing is I don't want to replace, I want to merge...


Lyle

From webmaster at cosmicperl.com  Wed Jul  9 22:49:08 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 09 Jul 2008 22:49:08 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>
References: <4872A52A.4030503@cosmicperl.com>
	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>
Message-ID: <48753254.3060709@cosmicperl.com>

Mark Hughes wrote:
> Hi Lyle,
> You could try the 'INSERT ... ON DUPLICATE KEY UPDATE' syntax
>
> http://dev.mysql.com/doc/refman/5.0/en/insert-on-duplicate.html
>
> Something like:
>
> insert into new_table  select * from old_table on duplicate key update
> new_table.totalamount = new_table.totalamount + old_table.totalamount;
>
> Or if old_table doesn't have any unique rows, or you don't want to add
> them to new_table just use a multiple table update
>
> update new_table  set new_table.totalamount = new_table.totalamount +
> old_table.totalamount where new_table.id = old_table.id

That sounds pretty interesting. I'll have a play about with it and see 
what I can get working... Thanks for the pointer.


Lyle


From psykx.out at googlemail.com  Fri Jul 11 07:36:44 2008
From: psykx.out at googlemail.com (psykx.out)
Date: Fri, 11 Jul 2008 07:36:44 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <487531F1.6040901@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com> <48746389.2070704@googlemail.com>
	<487531F1.6040901@cosmicperl.com>
Message-ID: <4876FF7C.7010000@googlemail.com>

Lyle wrote:
> psykx.out wrote:
>   
>> I'm not I understand perfectly, you can use 'replace into' instead of 
>> 'insert into' if you want to overwrite (edit) existing data. replace 
>> acts like insert if the is no existing record
>>     
>
> Thing is I don't want to replace, I want to merge...
>
>
> Lyle
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>   
Ah the beauty of replace, if you don't set all the fields in the table 
and it replaces another row it just leaves the old data there.

Max

From alex at alexfrancis.org.uk  Fri Jul 11 07:54:21 2008
From: alex at alexfrancis.org.uk (Alex Francis)
Date: Fri, 11 Jul 2008 07:54:21 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <48753254.3060709@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com>
	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>
	<48753254.3060709@cosmicperl.com>
Message-ID: <16afc9550807102354l7dc072fala670f62aad8f6416@mail.gmail.com>

On Wed, Jul 9, 2008 at 10:49 PM, Lyle <webmaster at cosmicperl.com> wrote:
> Mark Hughes wrote:
>> Hi Lyle,
>> You could try the 'INSERT ... ON DUPLICATE KEY UPDATE' syntax
>>
...
>
> That sounds pretty interesting. I'll have a play about with it and see
> what I can get working... Thanks for the pointer.
>

...

cool but watch out for bugs with this feature

e.g.
http://bugs.mysql.com/bug.php?id=29941

more:
http://bugs.mysql.com/search.php?search_for=ON+DUPLICATE+KEY+UPDATE&status[]=Active&severity=&limit=30&order_by=&cmd=display&phpver=&os=0&os_details=&bug_age=0&tags=&similar=&target=&defect_class=all&workaround_viability=all&impact=all&fix_risk=all&fix_effort=all

Alex

From jamie at linuxufo.com  Fri Jul 11 08:53:54 2008
From: jamie at linuxufo.com (Jamie Andrews)
Date: Fri, 11 Jul 2008 08:53:54 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <48753254.3060709@cosmicperl.com>
References: <4872A52A.4030503@cosmicperl.com>
	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>
	<48753254.3060709@cosmicperl.com>
Message-ID: <20080711075354.GA15819@linuxufo.com>

On Wed, Jul 09, 2008 at 10:49:08PM +0100, Lyle wrote:

I am a bit old skool but I'd restore the data into a different database
and then use a perl script to do the merge


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080711/b07dae7f/attachment.bin 

From webmaster at cosmicperl.com  Fri Jul 11 12:48:28 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 11 Jul 2008 12:48:28 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <16afc9550807102354l7dc072fala670f62aad8f6416@mail.gmail.com>
References: <4872A52A.4030503@cosmicperl.com>	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>	<48753254.3060709@cosmicperl.com>
	<16afc9550807102354l7dc072fala670f62aad8f6416@mail.gmail.com>
Message-ID: <4877488C.2060908@cosmicperl.com>

Alex Francis wrote:
> On Wed, Jul 9, 2008 at 10:49 PM, Lyle <webmaster at cosmicperl.com> wrote:
>   
>> Mark Hughes wrote:
>>     
>>> Hi Lyle,
>>> You could try the 'INSERT ... ON DUPLICATE KEY UPDATE' syntax
>>>
>>>       
> ...
>   
>> That sounds pretty interesting. I'll have a play about with it and see
>> what I can get working... Thanks for the pointer.
>>
>>     
>
> ...
>
> cool but watch out for bugs with this feature
>
> e.g.
> http://bugs.mysql.com/bug.php?id=29941
>
> more:
> http://bugs.mysql.com/search.php?search_for=ON+DUPLICATE+KEY+UPDATE&status[]=Active&severity=&limit=30&order_by=&cmd=display&phpver=&os=0&os_details=&bug_age=0&tags=&similar=&target=&defect_class=all&workaround_viability=all&impact=all&fix_risk=all&fix_effort=all
>   

Damn, that's pretty worrying, I've just written some important code 
using this feature :(


Lyle

From webmaster at cosmicperl.com  Fri Jul 11 12:59:42 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 11 Jul 2008 12:59:42 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <4876FF7C.7010000@googlemail.com>
References: <4872A52A.4030503@cosmicperl.com>
	<48746389.2070704@googlemail.com>	<487531F1.6040901@cosmicperl.com>
	<4876FF7C.7010000@googlemail.com>
Message-ID: <48774B2E.3080402@cosmicperl.com>

psykx.out wrote:
> Lyle wrote:
>   
>> psykx.out wrote:
>>   
>>     
>>> I'm not I understand perfectly, you can use 'replace into' instead of 
>>> 'insert into' if you want to overwrite (edit) existing data. replace 
>>> acts like insert if the is no existing record
>>>     
>>>       
>> Thing is I don't want to replace, I want to merge...
>>   
>>     
> Ah the beauty of replace, if you don't set all the fields in the table 
> and it replaces another row it just leaves the old data there

I didn't realize, thanks.


Lyle


From webmaster at cosmicperl.com  Fri Jul 11 13:11:32 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 11 Jul 2008 13:11:32 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <20080711075354.GA15819@linuxufo.com>
References: <4872A52A.4030503@cosmicperl.com>	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>	<48753254.3060709@cosmicperl.com>
	<20080711075354.GA15819@linuxufo.com>
Message-ID: <48774DF4.3020100@cosmicperl.com>

Jamie Andrews wrote:
> On Wed, Jul 09, 2008 at 10:49:08PM +0100, Lyle wrote:
>
> I am a bit old skool but I'd restore the data into a different database
> and then use a perl script to do the merge
>   

The database is a bit large. And it's part of a function that'll be run 
many times in the future with even bigger databases. Doing the merge 
with Perl would be far to slow.


Lyle


From webmaster at cosmicperl.com  Sat Jul 19 00:14:53 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Sat, 19 Jul 2008 00:14:53 +0100
Subject: [Bristolbathpm] Just saw that Virgin broadband advert again...
Message-ID: <488123ED.9070308@cosmicperl.com>

Hi All,
  Just saw that Virgin broadband advert again... Have you seen it? They 
say about BT having copper wire and them having fibre... Fair enough 
fibre is a lot better...

BUT...

And to me it's a big BUT...

They are offering 20Mb broadband, BT ADSL2+ is offering up to 24Mb 
broadband...

What's really bugging me, is, that as they have fibre, they should be 
offering at the very least, 100Mb broadband up and down. Why are they 
offering lower speeds than can be achieved over a twisted pair??? In my 
mind there is no excuse.

Is there some kind of fair competition law getting in the way or is 
Virgin just making some stupid decisions???


Lyle


From revford at blueyonder.co.uk  Sat Jul 19 09:44:10 2008
From: revford at blueyonder.co.uk (gav)
Date: Sat, 19 Jul 2008 09:44:10 +0100
Subject: [Bristolbathpm] Just saw that Virgin broadband advert again...
In-Reply-To: <488123ED.9070308@cosmicperl.com>
References: <488123ED.9070308@cosmicperl.com>
Message-ID: <20080719084410.GA15469@bluemidget.local>

On Sat, Jul 19, 2008 at 12:14:53AM +0100, Lyle wrote:
> What's really bugging me, is, that as they have fibre, they should be 
> offering at the very least, 100Mb broadband up and down. Why are they 
> offering lower speeds than can be achieved over a twisted pair??? In my 
> mind there is no excuse.
> 
> Is there some kind of fair competition law getting in the way or is 
> Virgin just making some stupid decisions???

At least in my case, Virgin run copper to the house.  Only the back end is 
fibre.

Also, they don't seem to actually have the capacity to provide the speeds they
already advertise.

Since they got the iPlayer shock, when people started to actually use the 
connections they pay for, they have taken to halving the connection speed if 
you download for more than an hour or so.

It's all in the terms and conditions, nothing sneaky, but it happens quite a 
lot if you watch telly online or download lots of large files like Linux disc
images.

If they were being more honest about it, they provide 10Mb regular speed, with
a 20Mb peak.  But that doesn't sound as impressive in an advert.

-- 
Gav Ford
revford at blueyonder.co.uk 
http://revford.pwp.blueyonder.co.uk 
I think we need to:  Attach the radar capacitor
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
Url : http://mailman.bristolbath.org/pipermail/bristolbathpm/attachments/20080719/06de8962/attachment.bin 

From webmaster at cosmicperl.com  Tue Jul 22 16:16:23 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 22 Jul 2008 16:16:23 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
Message-ID: <4885F9C7.9020603@cosmicperl.com>

Hi All,
  I've got a client who has a large database of clubs around the UK. 
They want to have a club locator pretty much the same as ASDAs store 
locator:-
http://www.asda.co.uk/corp/storelocator_frameset/storelocator_frameset.htm

Where you put in your town or postcode and it'll bring up the closest clubs.

I haven't done anything like this before, but I'm guessing there is some 
standard (ish) system that I get get to through Perl? Anyone have 
experience with this kind of thing, or know what I should be looking for?

A CPAN search has brought up 
http://search.cpan.org/~wross/Geo-Postcode-0.15/lib/Geo/Postcode.pm
But the more exact grid references I'm going to need to get the PAF from 
royal mail (over 2k per year), or possibly use an online system like 
http://www.alliescomputing.co.uk/ Web SOAP...

Hmm.. Maybe I should code a Web SOAP extension for Geo::Postcode?


Lyle


From robingarthedwards at googlemail.com  Tue Jul 22 16:25:19 2008
From: robingarthedwards at googlemail.com (Rob Edwards)
Date: Tue, 22 Jul 2008 16:25:19 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
In-Reply-To: <4885F9C7.9020603@cosmicperl.com>
References: <4885F9C7.9020603@cosmicperl.com>
Message-ID: <ded1116d0807220825y52f0b6d6udf6257064a380e53@mail.gmail.com>

http://developer.yahoo.com/addressbook/

Someone mentioned this the other day, maybe what you need.

Rob

2008/7/22 Lyle <webmaster at cosmicperl.com>:
> Hi All,
>  I've got a client who has a large database of clubs around the UK.
> They want to have a club locator pretty much the same as ASDAs store
> locator:-
> http://www.asda.co.uk/corp/storelocator_frameset/storelocator_frameset.htm
>
> Where you put in your town or postcode and it'll bring up the closest clubs.
>
> I haven't done anything like this before, but I'm guessing there is some
> standard (ish) system that I get get to through Perl? Anyone have
> experience with this kind of thing, or know what I should be looking for?
>
> A CPAN search has brought up
> http://search.cpan.org/~wross/Geo-Postcode-0.15/lib/Geo/Postcode.pm
> But the more exact grid references I'm going to need to get the PAF from
> royal mail (over 2k per year), or possibly use an online system like
> http://www.alliescomputing.co.uk/ Web SOAP...
>
> Hmm.. Maybe I should code a Web SOAP extension for Geo::Postcode?
>
>
> Lyle
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>

From webmaster at cosmicperl.com  Wed Jul 23 01:35:39 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 23 Jul 2008 01:35:39 +0100
Subject: [Bristolbathpm] Certification... your experiences
Message-ID: <48867CDB.1030203@cosmicperl.com>

Hi All,
  I'm looking into certification of programmers. I have no 
certifications myself, my only experiences are with what I've read about 
and when I sent my brother down to RedHat for their RHCT.

  I'm interested in what experiences people on this list have had with 
programmer certification in any language. What they thought was good/bad 
about it. Ideas on how they would improve it, etc.

Ultimately I'm interested in what you guys things about the idea of Perl 
Certification. I know it pops up at Perl Monks, use.perl, advocacy, etc, 
now and again. Strong arguments on both sides. Would be great to know 
who's for it, against it, and who's sitting on the fence :)


Lyle


From dave at dave.org.uk  Wed Jul 23 06:39:44 2008
From: dave at dave.org.uk (Dave Cross)
Date: Wed, 23 Jul 2008 06:39:44 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
In-Reply-To: <4885F9C7.9020603@cosmicperl.com>
References: <4885F9C7.9020603@cosmicperl.com>
Message-ID: <4886C420.9070300@dave.org.uk>

Lyle wrote:

> But the more exact grid references I'm going to need to get the PAF from 
> royal mail (over 2k per year),

You don't necessarily have to pay all that money.

   http://www.showusabetterway.co.uk/call/data.html#mail

Dave...

From webmaster at cosmicperl.com  Wed Jul 23 15:00:52 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Wed, 23 Jul 2008 15:00:52 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
In-Reply-To: <4886C420.9070300@dave.org.uk>
References: <4885F9C7.9020603@cosmicperl.com> <4886C420.9070300@dave.org.uk>
Message-ID: <48873994.8080308@cosmicperl.com>

Dave Cross wrote:
> Lyle wrote:
>
>   
>> But the more exact grid references I'm going to need to get the PAF from 
>> royal mail (over 2k per year),
>>     
>
> You don't necessarily have to pay all that money.
>
>    http://www.showusabetterway.co.uk/call/data.html#mail
>
> Dave...
>   

Thanks for the link. Looks very interesting, although I'd have to read 
through a lot more and check that getting a copy for what I need would 
be ok from a legal perspective... (probably should have worded that 
sentence better and broken it down)


Lyle

From psykx.out at googlemail.com  Thu Jul 24 01:51:35 2008
From: psykx.out at googlemail.com (psykx.out)
Date: Thu, 24 Jul 2008 01:51:35 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
In-Reply-To: <4885F9C7.9020603@cosmicperl.com>
References: <4885F9C7.9020603@cosmicperl.com>
Message-ID: <4887D217.6060104@googlemail.com>

Lyle wrote:
> Hi All,
>   I've got a client who has a large database of clubs around the UK. 
> They want to have a club locator pretty much the same as ASDAs store 
> locator:-
> http://www.asda.co.uk/corp/storelocator_frameset/storelocator_frameset.htm
>
> Where you put in your town or postcode and it'll bring up the closest clubs.
>
> I haven't done anything like this before, but I'm guessing there is some 
> standard (ish) system that I get get to through Perl? Anyone have 
> experience with this kind of thing, or know what I should be looking for?
>
> A CPAN search has brought up 
> http://search.cpan.org/~wross/Geo-Postcode-0.15/lib/Geo/Postcode.pm
> But the more exact grid references I'm going to need to get the PAF from 
> royal mail (over 2k per year), or possibly use an online system like 
> http://www.alliescomputing.co.uk/ Web SOAP...
>
> Hmm.. Maybe I should code a Web SOAP extension for Geo::Postcode?
>
>
> Lyle
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>   
We needed to do something similar at work, I think me bos used 
Pythagoras's therom and the paf data although I seem to remember 
something about a PHP function that did something with the PAF data

Max

p.s I've had a beer or two I respond again if I think this doesn't make 
enough sense or I remember more

From psykx.out at googlemail.com  Thu Jul 24 01:54:08 2008
From: psykx.out at googlemail.com (psykx.out)
Date: Thu, 24 Jul 2008 01:54:08 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <48867CDB.1030203@cosmicperl.com>
References: <48867CDB.1030203@cosmicperl.com>
Message-ID: <4887D2B0.4050205@googlemail.com>

Lyle wrote:
> Hi All,
>   I'm looking into certification of programmers. I have no 
> certifications myself, my only experiences are with what I've read about 
> and when I sent my brother down to RedHat for their RHCT.
>
>   I'm interested in what experiences people on this list have had with 
> programmer certification in any language. What they thought was good/bad 
> about it. Ideas on how they would improve it, etc.
>
> Ultimately I'm interested in what you guys things about the idea of Perl 
> Certification. I know it pops up at Perl Monks, use.perl, advocacy, etc, 
> now and again. Strong arguments on both sides. Would be great to know 
> who's for it, against it, and who's sitting on the fence :)
>
>
> Lyle
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>   
Basically as I see it now _all_ certs are bad. all of them period. if 
you can't explain to an interviewer why your would be good for the job 
in question you shouldn't have it. This come from somebody working in IT 
with no certs or qualifications.

Max

From paulm at paulm.com  Thu Jul 24 03:16:07 2008
From: paulm at paulm.com (Paul Makepeace)
Date: Wed, 23 Jul 2008 19:16:07 -0700
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4887D2B0.4050205@googlemail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
Message-ID: <c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>

On Wed, Jul 23, 2008 at 5:54 PM, psykx.out <psykx.out at googlemail.com> wrote:
> Lyle wrote:
>> Hi All,
>>   I'm looking into certification of programmers. I have no
>> certifications myself, my only experiences are with what I've read about
>> and when I sent my brother down to RedHat for their RHCT.
>>
>>   I'm interested in what experiences people on this list have had with
>> programmer certification in any language. What they thought was good/bad
>> about it. Ideas on how they would improve it, etc.
>>
>> Ultimately I'm interested in what you guys things about the idea of Perl
>> Certification. I know it pops up at Perl Monks, use.perl, advocacy, etc,
>> now and again. Strong arguments on both sides. Would be great to know
>> who's for it, against it, and who's sitting on the fence :)
>>
>>
>> Lyle
>>
>> _______________________________________________
>> BristolBathPM mailing list
>> BristolBathPM at bristolbath.org
>> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>>
> Basically as I see it now _all_ certs are bad. all of them period. if
> you can't explain to an interviewer why your would be good for the job
> in question you shouldn't have it. This come from somebody working in IT

A certificate can save you spending the time making that explanation.
Saying "I have a degree" saves a lot of time, on both sides.

When I went for a gig with the BBC they had me do a brainbench test. I
was quite impressed with it, it really felt to me like someone who
knew the language and what was important about it wrote it. I'm
curious about answers to this thread too (less though about the
meta-discussion about merits of certification).

P

> with no certs or qualifications.
>
> Max
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>

From webmaster at cosmicperl.com  Thu Jul 24 03:32:30 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Thu, 24 Jul 2008 03:32:30 +0100
Subject: [Bristolbathpm] Postcode Perl lookup...
In-Reply-To: <4887D217.6060104@googlemail.com>
References: <4885F9C7.9020603@cosmicperl.com> <4887D217.6060104@googlemail.com>
Message-ID: <4887E9BE.7020107@cosmicperl.com>

psykx.out wrote:
> We needed to do something similar at work, I think me bos used 
> Pythagoras's therom and the paf data although I seem to remember 
> something about a PHP function that did something with the PAF data
>
> Max
>
> p.s I've had a beer or two I respond again if I think this doesn't make 
> enough sense or I remember more
>   

Late night drinking and managing a post that makes sense without 
offending anyone. I'm impressed ;)


Lyle

From webmaster at cosmicperl.com  Thu Jul 24 03:54:05 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Thu, 24 Jul 2008 03:54:05 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4887D2B0.4050205@googlemail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
Message-ID: <4887EECD.7020806@cosmicperl.com>

psykx.out wrote:
> Basically as I see it now _all_ certs are bad. all of them period. if
> you can't explain to an interviewer why your would be good for the job 
> in question you shouldn't have it. This come from somebody working in IT 
> with no certs or qualifications.
>   

I'd agree from what I've seen... Apart from the Red Hat certificates. To 
be honest when I went to their testing centre and saw how they did it I 
was very impressed, and I could see why I had read so many posts saying 
that Red Hat certs are the ONLY viable Linux certs.


Lyle


From webmaster at cosmicperl.com  Thu Jul 24 04:26:11 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Thu, 24 Jul 2008 04:26:11 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
Message-ID: <4887F653.2090408@cosmicperl.com>

Paul Makepeace wrote:
> On Wed, Jul 23, 2008 at 5:54 PM, psykx.out <psykx.out at googlemail.com> wrote:
>   
>> Basically as I see it now _all_ certs are bad. all of them period. if
>> you can't explain to an interviewer why your would be good for the job
>> in question you shouldn't have it. This come from somebody working in IT
>>     
>
> A certificate can save you spending the time making that explanation.
> Saying "I have a degree" saves a lot of time, on both sides.
>
> When I went for a gig with the BBC they had me do a brainbench test. I
> was quite impressed with it, it really felt to me like someone who
> knew the language and what was important about it wrote it. I'm
> curious about answers to this thread too (less though about the
> meta-discussion about merits of certification).
>
> P
>   

That's interesting. From Perl Cert posts I've read on LPM and Perl Monks 
a lot of people don't rate Brainbench tests. The main argument being 
that the tests are based on memory of functions and their parameters, as 
opposed to whether you can actually write code. In fairness if you can 
remember a lot of functions and their parameters without having to look 
them up, you'll potentially be a faster programmer (or at least spend 
less time looking up functions and their parameters).
  Although in fairness BrainBench are about the only ones (I did once 
come across another) that offer some kind of Perl Certificate. This 
could be clear evidence that the BBC at least prefer to deal with 
certified programmers.


Lyle


From paulm at paulm.com  Thu Jul 24 06:33:15 2008
From: paulm at paulm.com (Paul Makepeace)
Date: Wed, 23 Jul 2008 22:33:15 -0700
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4887F653.2090408@cosmicperl.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
Message-ID: <c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>

On Wed, Jul 23, 2008 at 8:26 PM, Lyle <webmaster at cosmicperl.com> wrote:
> Paul Makepeace wrote:
>> On Wed, Jul 23, 2008 at 5:54 PM, psykx.out <psykx.out at googlemail.com> wrote:
>>
>>> Basically as I see it now _all_ certs are bad. all of them period. if
>>> you can't explain to an interviewer why your would be good for the job
>>> in question you shouldn't have it. This come from somebody working in IT
>>>
>>
>> A certificate can save you spending the time making that explanation.
>> Saying "I have a degree" saves a lot of time, on both sides.
>>
>> When I went for a gig with the BBC they had me do a brainbench test. I
>> was quite impressed with it, it really felt to me like someone who
>> knew the language and what was important about it wrote it. I'm
>> curious about answers to this thread too (less though about the
>> meta-discussion about merits of certification).
>>
>> P
>>
>
> That's interesting. From Perl Cert posts I've read on LPM and Perl Monks
> a lot of people don't rate Brainbench tests. The main argument being
> that the tests are based on memory of functions and their parameters, as
> opposed to whether you can actually write code.

I haven't read the posts so was wondering if they had an opinion on
how you would test whether someone could write code? I can see that's
kinda testable in an interview but doing it in a certifiable way seems
tough to me.

> In fairness if you can
> remember a lot of functions and their parameters without having to look
> them up, you'll potentially be a faster programmer (or at least spend
> less time looking up functions and their parameters).
>  Although in fairness BrainBench are about the only ones (I did once
> come across another) that offer some kind of Perl Certificate. This
> could be clear evidence that the BBC at least prefer to deal with
> certified programmers.

I got the impression working there the BBC is a big diverse
organisation so it's quite possible that that was just the group I
ended up working in. (Compared with say Google where the hiring &
interview process is controlled & coordinated across the whole
company).

P

>
>
> Lyle
>
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>

From david at cantrell.org.uk  Thu Jul 24 14:49:41 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Thu, 24 Jul 2008 14:49:41 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
Message-ID: <20080724134941.GQ9038@bytemark.barnyard.co.uk>

On Wed, Jul 23, 2008 at 07:16:07PM -0700, Paul Makepeace wrote:

> A certificate can save you spending the time making that explanation.
> Saying "I have a degree" saves a lot of time, on both sides.

We all know of people who have degrees and are utterly clueless.  I
expect most of us have, unfortunately, had to work with such people or
clean up after them.

"I have a degree" doesn't mean anything, and "I have a degree in
computer science" doubly so.  About the only degree that I'll pay any
attention to is something like theology.  The sort of twisty thinking
that requires is dead useful for debugging and using third-party
libraries.

-- 
David Cantrell | Hero of the Information Age

comparative and superlative explained:

<Huhn> worse, worser, worsest, worsted, wasted

From david at cantrell.org.uk  Thu Jul 24 15:05:35 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Thu, 24 Jul 2008 15:05:35 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
Message-ID: <20080724140535.GR9038@bytemark.barnyard.co.uk>

On Wed, Jul 23, 2008 at 10:33:15PM -0700, Paul Makepeace wrote:
> On Wed, Jul 23, 2008 at 8:26 PM, Lyle <webmaster at cosmicperl.com> wrote:
> > That's interesting. From Perl Cert posts I've read on LPM and Perl Monks
> > a lot of people don't rate Brainbench tests. The main argument being
> > that the tests are based on memory of functions and their parameters, as
> > opposed to whether you can actually write code.
> I haven't read the posts so was wondering if they had an opinion on
> how you would test whether someone could write code? I can see that's
> kinda testable in an interview but doing it in a certifiable way seems
> tough to me.

Here at $work we test people by having them come in for a programming
test.  We give them an hour and a half with perl and an interwebnet
connection to solve as much as they can of a fairly open-ended problem.
The problem tests knowledge of algorithms and data structures, and the
really good people get enough done that they can start on optimising
their code.

In the interview we go through the code together, discussing other
approaches they might have taken, why they chose particularly methods,
and also talk about testing and documentation (which we don't expect
them to do in the test itself).

Yes, it's hard to certify - although don't universities and schools
manage to do it in computer science courses? yes they do, but to do so
is expensive - but we don't *need* that.  We need people who can
convince me and my colleagues that they're competent.

-- 
David Cantrell | Godless Liberal Elitist

You can't spell "slaughter" without "laughter"

From webmaster at cosmicperl.com  Thu Jul 24 15:46:01 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Thu, 24 Jul 2008 15:46:01 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <20080724140535.GR9038@bytemark.barnyard.co.uk>
References: <48867CDB.1030203@cosmicperl.com>
	<4887D2B0.4050205@googlemail.com>	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>	<4887F653.2090408@cosmicperl.com>	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
	<20080724140535.GR9038@bytemark.barnyard.co.uk>
Message-ID: <488895A9.6060406@cosmicperl.com>

David Cantrell wrote:
> On Wed, Jul 23, 2008 at 10:33:15PM -0700, Paul Makepeace wrote:
>   
>> On Wed, Jul 23, 2008 at 8:26 PM, Lyle <webmaster at cosmicperl.com> wrote:
>>     
>>> That's interesting. From Perl Cert posts I've read on LPM and Perl Monks
>>> a lot of people don't rate Brainbench tests. The main argument being
>>> that the tests are based on memory of functions and their parameters, as
>>> opposed to whether you can actually write code.
>>>       
>> I haven't read the posts so was wondering if they had an opinion on
>> how you would test whether someone could write code? I can see that's
>> kinda testable in an interview but doing it in a certifiable way seems
>> tough to me.
>>     
>
> Here at $work we test people by having them come in for a programming
> test.  We give them an hour and a half with perl and an interwebnet
> connection to solve as much as they can of a fairly open-ended problem.
> The problem tests knowledge of algorithms and data structures, and the
> really good people get enough done that they can start on optimising
> their code.
>   

Do you test everyone who applies?
If not what criteria do you use to choose which ones you test?
How many do you generally end up testing?
How long would you say on average this takes to fill one position? (how 
long in hours of your time, not how long from job advert)


Sorry about all the questions :)


Lyle


From david at cantrell.org.uk  Thu Jul 24 18:06:24 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Thu, 24 Jul 2008 18:06:24 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <488895A9.6060406@cosmicperl.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
	<20080724140535.GR9038@bytemark.barnyard.co.uk>
	<488895A9.6060406@cosmicperl.com>
Message-ID: <20080724170623.GA11604@bytemark.barnyard.co.uk>

On Thu, Jul 24, 2008 at 03:46:01PM +0100, Lyle wrote:
> David Cantrell wrote:
> > Here at $work we test people by having them come in for a programming
> > test ...
> Do you test everyone who applies?

No, only those who look like they have suitable experience and
knowledge.  The test is really a way of weeding out people who lie on
their CV.  They're also designed to seperate the merely good from the
excellent - an hour and a half isn't enough time to completely solve any
of the tasks we set, but we can get a good idea of how splendid someone
is by how far they get.

> If not what criteria do you use to choose which ones you test?

Experience, length of time in their various jobs, the usual.  We *don't*
care about their education, unless it's really interesting.  Like if
they learned to speak Zulu, or are an authority on sword-smithing, or
have a doctorate in theology from the Vatican university.

> How many do you generally end up testing?

Between a quarter and a half of those who apply.  For the last opening
we had, I think we had four people in, out of about ten applicants.  We
test everyone who we invite to interview for programming jobs.

> How long would you say on average this takes to fill one position? (how 
> long in hours of your time, not how long from job advert)

About half an hour to set a machine up for people to use in the test,
half an hour after each candidate has dunnit to discuss their code, and
wipe the user account ready for the next victim.  All of that is, of
course, on top of the time taken to read CVs, make paper aeroplanes out
of the rubbish ones, and do the rest of the interview.

We find that the small investment in time of administering a programming
test is well worth it.  It's weeded out a lot of people whose CVs didn't
scream "idiot" but who had difficulty with really simple concepts like
how to open a file (yes, really), and we've never had a complete numpty
get through that part of the interview process undetected.

Incidentally, we found that we had to use simpler tests for Java people,
because it takes longer to write the same functionality in Java.  Also
that idiots are far more likely to apply for Java jobs than perl jobs.
The guy who couldn't open files was an experienced Java programmer
according to his CV.

And it's tests, plural.  We've found that when people come from a
recruitment agency, the pimp phones them immediately after the interview
to talk to them, and finds out what the test was.  They then tell their
next candidate what it is, so they have longer to think about the
problem. So now we have several different ones.

-- 
David Cantrell | Official London Perl Mongers Bad Influence

    Immigration: making Britain great since AD43

From david at cantrell.org.uk  Thu Jul 24 18:15:12 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Thu, 24 Jul 2008 18:15:12 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <20080724170623.GA11604@bytemark.barnyard.co.uk>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
	<20080724140535.GR9038@bytemark.barnyard.co.uk>
	<488895A9.6060406@cosmicperl.com>
	<20080724170623.GA11604@bytemark.barnyard.co.uk>
Message-ID: <20080724171512.GC11604@bytemark.barnyard.co.uk>

On Thu, Jul 24, 2008 at 06:06:24PM +0100, David Cantrell wrote:
> On Thu, Jul 24, 2008 at 03:46:01PM +0100, Lyle wrote:
> > How long would you say on average this takes to fill one position? (how 
> > long in hours of your time, not how long from job advert)
> About half an hour to set a machine up for people to use in the test,
> half an hour after each candidate has dunnit to discuss their code, and
> wipe the user account ready for the next victim.

Oh, plus an hour and a half for each of two or three of us when we
create a new test - need to do it ourselves so we can have a good idea
of how hard it is and what the pitfalls might be.  But if we use a test
ten or fifteen times, that time does't really count for much.

-- 
David Cantrell | Nth greatest programmer in the world

Never attribute to malice that which can be explained by stupidity
    -- Hanlon's Razor

Stupidity maintained long enough is a form of malice
    -- Richard Bos's corollary

From paulm at paulm.com  Thu Jul 24 18:18:58 2008
From: paulm at paulm.com (Paul Makepeace)
Date: Thu, 24 Jul 2008 10:18:58 -0700
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <20080724170623.GA11604@bytemark.barnyard.co.uk>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
	<20080724140535.GR9038@bytemark.barnyard.co.uk>
	<488895A9.6060406@cosmicperl.com>
	<20080724170623.GA11604@bytemark.barnyard.co.uk>
Message-ID: <c37577dd0807241018y7c5a91e6y1f54dcdfe39ae92f@mail.gmail.com>

On Thu, Jul 24, 2008 at 10:06 AM, David Cantrell <david at cantrell.org.uk> wrote:

> And it's tests, plural.  We've found that when people come from a
> recruitment agency, the pimp phones them immediately after the interview
> to talk to them, and finds out what the test was.  They then tell their
> next candidate what it is, so they have longer to think about the
> problem. So now we have several different ones.

That's really bad. Do you tell the candidate that they're bound to
keep quiet about the contents of the interview?

The other trick is to ask questions that even with having read it
online you can go off in different directions and reveal the
candidate's true knowledge/wit pretty effectively.

P

>
> --
> David Cantrell | Official London Perl Mongers Bad Influence
>
>    Immigration: making Britain great since AD43
> _______________________________________________
> BristolBathPM mailing list
> BristolBathPM at bristolbath.org
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>

From david at cantrell.org.uk  Thu Jul 24 19:18:22 2008
From: david at cantrell.org.uk (David Cantrell)
Date: Thu, 24 Jul 2008 19:18:22 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <c37577dd0807241018y7c5a91e6y1f54dcdfe39ae92f@mail.gmail.com>
References: <48867CDB.1030203@cosmicperl.com> <4887D2B0.4050205@googlemail.com>
	<c37577dd0807231916n3b092e96o357045e028657972@mail.gmail.com>
	<4887F653.2090408@cosmicperl.com>
	<c37577dd0807232233v5b2bc628t9703742289b6cbcb@mail.gmail.com>
	<20080724140535.GR9038@bytemark.barnyard.co.uk>
	<488895A9.6060406@cosmicperl.com>
	<20080724170623.GA11604@bytemark.barnyard.co.uk>
	<c37577dd0807241018y7c5a91e6y1f54dcdfe39ae92f@mail.gmail.com>
Message-ID: <20080724181822.GA4635@bytemark.barnyard.co.uk>

On Thu, Jul 24, 2008 at 10:18:58AM -0700, Paul Makepeace wrote:
> On Thu, Jul 24, 2008 at 10:06 AM, David Cantrell <david at cantrell.org.uk> wrote:
> > And it's tests, plural.  We've found that when people come from a
> > recruitment agency, the pimp phones them immediately after the interview
> > to talk to them, and finds out what the test was.  They then tell their
> > next candidate what it is, so they have longer to think about the
> > problem. So now we have several different ones.
> That's really bad. Do you tell the candidate that they're bound to
> keep quiet about the contents of the interview?

We didn't think to do it until after the pimps started being dicks.

-- 
David Cantrell | top google result for "internet beard fetish club"

  I don't do .INI, .BAT, or .SYS files.  I don't assign apps to files.
  I don't configure peripherals or networks before using them. I have
  a computer to do all that. I have a Macintosh, not a hobby.
        -- Fritz Anderson

From mjr at phonecoop.coop  Fri Jul 25 11:27:36 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Fri, 25 Jul 2008 11:27:36 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <48867CDB.1030203@cosmicperl.com>
References: <48867CDB.1030203@cosmicperl.com>
Message-ID: <4889aa98.awQVfA4XCds5F5bs%mjr@phonecoop.coop>

Lyle <webmaster at cosmicperl.com> wrote:
>   I'm interested in what experiences people on this list have had with 
> programmer certification in any language. What they thought was good/bad 
> about it. Ideas on how they would improve it, etc.

My main experiences are:-

1. as a teacher and advisory board member, seeing the discrepency
between modern mainstream education with its portfolio-based
assessment programmes, and current certification systems with their
(usually paper) short-term tests;

2. as a business owner, with people who have passed (fluked?) a
programmer certificate and use it to try to convince us they can
program when they can't.

I think the strength in the mainstream education certificates comes
from the human assessment, which seems to be a feature that many
programmer certifications seem to try to remove, trying to make the
assessments as automated as they can.

There is a danger in assessment that is too human-based, in that some
certificates become essentially "old boy network" ones, where it's
much easier to get them if you move in the right circles.  Automated
assessments are verifiable, but they can be too crammable I guess.


> Ultimately I'm interested in what you guys things about the idea of Perl 
> Certification. I know it pops up at Perl Monks, use.perl, advocacy, etc, 
> now and again. Strong arguments on both sides. Would be great to know 
> who's for it, against it, and who's sitting on the fence :)

Strongly against.  Do a good demonstration instead.  There are tons of
free software and voluntary-sector projects other there which would
appreciate the help.  References can also be good, but sharing the
code is better, at least from where I'm sat in the third sector.  I
guess some private sector business might prefer people to show they
can keep code private!


"Paul Makepeace" <paulm at paulm.com> wrote: [...]
> A certificate can save you spending the time making that explanation.
> Saying "I have a degree" saves a lot of time, on both sides. [...]

Only if the hiring side wants to risk a horrendous mistake.  If you're
lucky, a good degree will give someone the basics of algorithms, data
structures, time management and research skills.  It won't tell you
whether they can actually do the work you've got.


Lyle <webmaster at cosmicperl.com> wrote:
> was very impressed, and I could see why I had read so many posts saying 
> that Red Hat certs are the ONLY viable Linux certs.

I've heard OK things about the Red Hat certs but I've not seen
anything which suggests that they're viable GNU/Linux certs rather
than just Red Hat certs.


David Cantrell <david at cantrell.org.uk> wrote: [...]
> We didn't think to do it until after the pimps started being dicks.

I think you need new pimps^Wrecruiters.

Regards,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From webmaster at cosmicperl.com  Fri Jul 25 13:48:03 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 25 Jul 2008 13:48:03 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4889aa98.awQVfA4XCds5F5bs%mjr@phonecoop.coop>
References: <48867CDB.1030203@cosmicperl.com>
	<4889aa98.awQVfA4XCds5F5bs%mjr@phonecoop.coop>
Message-ID: <4889CB83.6070300@cosmicperl.com>

MJ Ray wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>   
>> was very impressed, and I could see why I had read so many posts saying 
>> that Red Hat certs are the ONLY viable Linux certs.
>>     
>
> I've heard OK things about the Red Hat certs but I've not seen
> anything which suggests that they're viable GNU/Linux certs rather
> than just Red Hat certs.
>   

The other Linux certs I've seen such as LPI seem to be multiple choice.

The Red Hat certs are totally human based, no paper work (at least not 
for the RHCE and I guess higher up certs as well).
Each test candidate is given a machine that isn't working. They are told 
to fix it given minimum information about the problems. They have to 
score 100% i.e. have the machine fully working in that section to pass. 
If they pass that section then they are given a list of things they need 
to setup / configure in the next section. The candidates machines are 
human checked for errors and whether the different parts are setup/working.



Lyle


From mjr at phonecoop.coop  Fri Jul 25 14:11:38 2008
From: mjr at phonecoop.coop (MJ Ray)
Date: Fri, 25 Jul 2008 14:11:38 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4889CB83.6070300@cosmicperl.com>
References: <48867CDB.1030203@cosmicperl.com>
	<4889aa98.awQVfA4XCds5F5bs%mjr@phonecoop.coop>
	<4889CB83.6070300@cosmicperl.com>
Message-ID: <4889d10a.aM/NvyhJKzWz5+DR%mjr@phonecoop.coop>

Lyle <webmaster at cosmicperl.com> wrote:
> MJ Ray wrote:
> > I've heard OK things about the Red Hat certs but I've not seen
> > anything which suggests that they're viable GNU/Linux certs rather
> > than just Red Hat certs.
>
> The other Linux certs I've seen such as LPI seem to be multiple choice.

Indeed, and LPI also had a bit too much NIH and they support
Perens-flamed former swpat-supporters OpenForumEurope...

> The Red Hat certs [...] The candidates machines are 
> human checked for errors and whether the different parts are setup/working.

Sure, but I've been told that setting something up according to, say,
File Hierarchy Standard, can get you marked down for not doing it the
Red Hat way.  They're Red Hat certs, after all, aren't they?  Which
isn't all that great for real heterogenous sysadminning.

Regards,
-- 
MJ Ray (slef)
Webmaster for hire, statistician and online shop builder for a small
worker cooperative http://www.ttllp.co.uk/ http://mjr.towers.org.uk/
(Notice http://mjr.towers.org.uk/email.html) tel:+44-844-4437-237

From webmaster at cosmicperl.com  Fri Jul 25 17:10:30 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Fri, 25 Jul 2008 17:10:30 +0100
Subject: [Bristolbathpm] Certification... your experiences
In-Reply-To: <4889d10a.aM/NvyhJKzWz5+DR%mjr@phonecoop.coop>
References: <48867CDB.1030203@cosmicperl.com>	<4889aa98.awQVfA4XCds5F5bs%mjr@phonecoop.coop>	<4889CB83.6070300@cosmicperl.com>
	<4889d10a.aM/NvyhJKzWz5+DR%mjr@phonecoop.coop>
Message-ID: <4889FAF6.8050901@cosmicperl.com>

MJ Ray wrote:
> Lyle <webmaster at cosmicperl.com> wrote:
>   
>> The Red Hat certs [...] The candidates machines are 
>> human checked for errors and whether the different parts are setup/working.
>>     
>
> Sure, but I've been told that setting something up according to, say,
> File Hierarchy Standard, can get you marked down for not doing it the
> Red Hat way.  They're Red Hat certs, after all, aren't they?  Which
> isn't all that great for real heterogenous sysadminning.
>   

I agree they are aimed for just the RedHat platform and aren't supposed 
to be a general Linux cert.

What I'm impressed with is their testing methodology. To me it seems 
clearly a much better way of doing it.

If I were hiring a sysadmin to work on RedHat based machines (including 
CentOS, etc) and they had the RHCE certificate, I would feel no need to 
test them on the skills covered by the RHCE certificate (unless it was 
very old).

Obviously I'd still test them on some job specific tasks relating to 
what we need exactly.

But the cert would make my tests shorter, and help me choose potential 
candidates for the position easier. Obviously candidates for the 
position without the RedHat cert would still be viable, but I'd test 
that they had the skills on the RedHat cert to make sure they were 
capable. Which would take up more of my time.


I'm Strongly against Bad Perl Certification.

I'm Strongly for Good Perl Certification.


Lyle


From webmaster at cosmicperl.com  Tue Jul 29 04:02:45 2008
From: webmaster at cosmicperl.com (Lyle)
Date: Tue, 29 Jul 2008 04:02:45 +0100
Subject: [Bristolbathpm] Mysqldump restoring and updating
In-Reply-To: <16afc9550807102354l7dc072fala670f62aad8f6416@mail.gmail.com>
References: <4872A52A.4030503@cosmicperl.com>	<ba3a03420807090447v5bac1634v1b92055dbeb83b92@mail.gmail.com>	<48753254.3060709@cosmicperl.com>
	<16afc9550807102354l7dc072fala670f62aad8f6416@mail.gmail.com>
Message-ID: <488E8855.6020404@cosmicperl.com>

Alex Francis wrote:
>> Mark Hughes wrote:
>>     
>>> Hi Lyle,
>>> You could try the 'INSERT ... ON DUPLICATE KEY UPDATE' syntax
>>>
>>>       
> cool but watch out for bugs with this feature
>
> more:
> http://bugs.mysql.com/search.php?search_for=ON+DUPLICATE+KEY+UPDATE&status[]=Active&severity=&limit=30&order_by=&cmd=display&phpver=&os=0&os_details=&bug_age=0&tags=&similar=&target=&defect_class=all&workaround_viability=all&impact=all&fix_risk=all&fix_effort=all
>   

Just realized that this is MySQL specific, which makes future ports to 
MSSQL impossible.

Looks like I'm going to have to do a Transaction instead... Which should 
be fun as I haven't done any before :)


Lyle