Re: mutt

[Matt Sergeant]

Peter Corlett 
July 25, 2011 5:55 AM


On Sun, Jul 24, 2011 at 11:00:13PM +0100, James Laver wrote:
[...]

It's quite shameful for RIM, given their devices are basically designed as
email terminals with a few other features added on as an afterthought.

The iOS mail client is best described as "adequate". It's arguably better
than Outlook, which seems to be the standard MUA these days.


Not even close - that's just selection bias.

http://www.campaignmonitor.com/stats/email-clients/ (caveats apply, etc).



__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Someone needs to take jwz aside...

[Matt Sergeant]

Simon Wistow wrote:

On Wed, Jun 08, 2011 at 10:57:56AM -0400, Matt Sergeant said:
   

I'm actually liking it more than CPAN for publishing and installing stuff.
The only weak area is lack of search.cpan.org.
 


My problem with npm is that it either tries to install stuff in some
random directory in my home directory or suggests, with a straight face,
that I chown -R $USER /usr/local
   


I think that has changed now with 1.0. Every install is into the current 
directory now.


I think you'll find it is pretty close to bundler now. Though the 
documentation is still lacking.


One really nice thing I like is if you're working on a project, you put 
your dependencies into the "package.json" in your project directory, and 
then just type "npm install" and it pulls in those dependencies 
automatically and makes them available to that project (and only that 
project).


Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Someone needs to take jwz aside...

[Matt Sergeant]

David Cantrell wrote:

It's the lack of a CPAN-a-like for any other language that keeps me
coming back to perl.

Of course, it's possible that the Comprehensive Python Archive Network
or similar for ruby/javascript/java/C/whatever does exist but I just
can't find it.  But then, if I can't find it, it's not much use.

http://npmjs.org/

I'm actually liking it more than CPAN for publishing and installing 
stuff. The only weak area is lack of search.cpan.org.


Note that this is for Node.js stuff only - not Javascript in general.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Someone needs to take jwz aside...

[Matt Sergeant]

Hakim Cassimally wrote:

While Javascript-the-language is lovely (as you say, better in some
respects, worse in others, than Perl), that's only one part of the
story.  I've not followed Javascript-the-platform that closely (i.e.
anything much beyond jQuery) - what's your experience been like,
working with Node and other libraries?


It's "ok". The standard library shipping with node is lacking in areas 
(there's no flock(), there's no getopts parser, and a bunch of other 
things), and right now it feels like it is stagnating while they get in 
true Windows support, but I've been surprised by the breadth of things 
available on npm (the equivalent of cpan).



Perl->Javascript is a really interesting migration path I'd not
considered, and I'm not sure "it's faster!" would convince me on its
own -- we all know there are faster languages than Perl.  But... JS
does have a significant advantages over, say, Perl->Haskell, as
Javascript is so widespread and therefore has many(devs, projects,
jobs).


OK let me put it another way than just "faster" - it's the fastest 
option for dynamic languages (with the possible exception of Lua-JIT) 
that has a sane (ALGOL-style) syntax, while still having a large user base.


My thinking behind Haraka is that a lot of web sites need a mail server 
with custom functionality, and their web site coders know Javascript, so 
why not provide them with an option to do stuff in JS for email.


Also the other huge thing for me is I've been writing async code in Perl 
now for years, and you always come across fighting with some library 
because it's got a blocking API so either you can't use it, or you take 
a risk and block the event loop while it runs (the same is true for 
Twisted or the Ruby one). That just doesn't happen in Node - everything 
is async.


Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Someone needs to take jwz aside...

[Matt Sergeant]

Simon Cozens wrote:

On 02/06/2011 21:50, gvim wrote:
   

>  Considering the amount of development you've done on Perl web frameworks over
>  the years isn't this tantamount to having given up on Perl, at least for web
>  development?
 


Yes and no. I've moved from being more of a developer to being more of a user.
Perl is a fantastic language for developers. It has a great culture for
developers. We're all brilliant at producing tools which other developers can
pick up and do really great stuff with. Perl is wonderful if I want to write
my own web framework, or construct my own CMS on top of one of the hundreds of
Perl web frameworks which already exist.
   


As someone else who has written a bunch of popular perl stuff over the 
years, I'll chime in here too - I write a lot less open source stuff 
these days, but when I do I'm looking much more to JavaScript. The 
language is actually about as good as Perl (some areas better, some 
worse), but the implementation, the interpreters, are just WAY faster.


https://github.com/baudehlo/Haraka

Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Speed v Version

[Matt Sergeant]

Dave Hodgkinson wrote:

My BBC sandbox is sane at least:

$ uname -p
x86_64
   


Shouldn't a BBC report "6502"? ;-)

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Xcode 4.0.2 and XS modules

[Matt Sergeant]

Nicholas Clark wrote:

Would they care more if they got lots of polite bug reports from registered
developers who care about Apple's SNAFU, encouraging Apple to re-instate the
PPC assembler for XCode 4? Is XCode 4*supposed*  to support the PPC-enabled
OS X versions? Or is it "Lion only"?
   

It's supposed to remove PPC support in preparation for Lion.


Also, did this bug start with XCode 4.0.whatever-the-first-one-is, or was
the bug added/the functionality removed in a "point" release?


I downloaded a beta version (I'm a paid-up ADC member) and it was there.

I've filed a report with apple. Bug Problem ID: 9340360

I noted on the bug report that you can "fix" it. Just edit -arch ppc out 
of Config_heavy.pl


Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Someone needs to take jwz aside...

[Matt Sergeant]

Abigail wrote:

On Wed, Apr 20, 2011 at 06:43:48PM +0200, Lars Thegler wrote:
   

On Wed, Apr 20, 2011 at 10:40 AM, Dave Hodgkinson  wrote:
 

http://www.jwz.org/blog/2011/04/a-badge-for-the-software-industrys-failures/

Or does he have a point?
   

No, code reuse is a *good* thing.
 


Sometimes.

But I've seen so much code that's needlessly convulated just so it can
just some code (or worse, that the code can be reused), that I take the
dogma "code reuse" with more salt than the yearly recommended dosis.
   
Especially when the code reuse in this case is JUST to get access to 
strptime.


Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Recommendation for simple Web Frameworks

[Matt Sergeant]

Eden Cardim wrote:

"Simon" == Simon Wistow  writes:


 Simon>  In short - I don't really need the CRUD stuff from a
 Simon>  framework, I really just need the url based dispatch. I
 Simon>  played around with Catalyst (which I'm familiar with from 6A)
 Simon>  but it felt like it was a bit of a sledgehammer and that it
 Simon>  was (not unreasonably) tied to an ORM.

You got the wrong impression, Catalyst isn't tied to anything, it's just
a plugglable/configurable http dispatcher, it's as simple as any of the
other perl web frameworks in that regard. It does have a larger user
base for testing/contribution/support, I'd say that's the main
advantage.


But the dependencies list *is* much larger for Catalyst.

http://deps.cpantesters.org/?module=Dancer;perl=latest

vs

http://deps.cpantesters.org/?module=Catalyst::Runtime;perl=latest

Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Sanity check - DateTime on Perl 5.12.1

[Matt Sergeant]

Toby Wintermute wrote:

Hi,
I wondered if anyone else running Perl 5.12.1 (and 12.0 would be
interesting too) could quickly check if they can build DateTime 0.5x
and pass the unit tests?

I have them failing on two machines, but they're very similar and I am
worried I might have screwed something up elsewhere, since I have been
playing with a lot of versions of Perl on said machines.
   


You know that Perl 5.12 ships with a pretty awesome date/time library, 
right? ;-)


(yeah I know it doesn't do everything DateTime does, it's a good 
lightweight alternative)


Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Founding a Perlmongers group

[Matt Sergeant]

James Laver wrote:

As I shall shortly be leaving london for somewhere with no PM group, I
thought it might be nice to create one.

How do I go about it?
   


http://www.pm.org/start/index.html

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: [Fwd: Betonmarkets CTO position]

[Matt Sergeant]

Paul Orrock wrote:
Things have moved on a lot since then. Firstly these two companies are 
not in any way related and secondly no reputable online gaming company 
offers accounts to US customers, thus avoid the need for anyone 
working for them to get arrested mid-flight. 


It's a shame they don't wise up and see eTrade as online gambling too.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: 2 depend or not 2 depend

[Matt Sergeant]

Nicholas Clark wrote:

It does if you have a second machine to test on.
It doesn't if you have a shared development server, and the installed packages
are common to all developers.


Then the owners of those boxes need to learn about xen. And fast.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: 2 depend or not 2 depend

[Matt Sergeant]

Ash Berlin wrote:

>  Use your OS's package management system.
 


Which is pretty much guaranteed to not have the exact versions they currently 
have installed if they've been using `cpan` et al. to install it .


I don't mean get them from the OS distributor. I mean build RPMs (or 
debs or whatever) and maintain them separately, including an SRPM 
archive if you ever need to rebuild stuff.


Use a tool, like cpan2rpm.

Matt.


__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: 2 depend or not 2 depend

[Matt Sergeant]

Paulo Edgar Castro wrote:
It's very likely that I might be missing some obvious knowledge/tricks 
here.
Would there be a cleverer way to accomplish this ? 

Use your OS's package management system.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Domain acquisition

[Matt Sergeant]

Jurgen Pletinckx wrote:

Hm. But that really only holds for domains you're actually using, or have
plans for, right? Can I actually find out which other domains the
proprietors hold? A reverse whois, so to say.
There are some services which can do this - they do it by downloading 
the .com zone every day and linking back to the nameservers. Doesn't 
always work (e.g. when using godaddy's or some other public DNS servers).


I might have an account on such a service if you need help.

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Perl Christmas Quiz 2009

[Matt Sergeant]

On Tue, 01 Dec 2009 08:47:55 -0800, Randal L. Schwartz wrote:
> Lemme guess.  You did this:
> 
> $ perl
> ... type program in here ...
> ^D (control D)
> 
> The "D" is from your control D.
> 
> Common misconception.

Yes. Though oddly enough it doesn't show up in the same terminal when 
ssh'd into a Linux box. I'd like to know the reason why that is.

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Perl Christmas Quiz 2009

[Matt Sergeant]

On Tue, 1 Dec 2009 11:23:09 -0500, Matt Sergeant wrote:
> On Mon, 30 Nov 2009 18:24:12 +, Chris Jack wrote:
>> 1) Without running it to check, what does the following program output?
>> 
>> 
>> my %a = (3,2,1,0);
>> 
>> 
>> for my $b (sort values %a) {
>> $b += 4;
>> }
>> 
>> 
>> print $a{1} . "\n";
> 
> Bizarrely enough, on both my Snow Leopard machines (default perl 
> install) this outputs: 4D
> 
> Nice bug there.

Ah. It's the D from ^D. Shitty terminal.

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Perl Christmas Quiz 2009

[Matt Sergeant]

On Mon, 30 Nov 2009 18:24:12 +, Chris Jack wrote:
> 1) Without running it to check, what does the following program output?
> 
> 
> my %a = (3,2,1,0);
> 
> 
> for my $b (sort values %a) {
> $b += 4;
> }
> 
> 
> print $a{1} . "\n";

Bizarrely enough, on both my Snow Leopard machines (default perl 
install) this outputs: 4D

Nice bug there.

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: keyboards/RSI/switching costs (was Looking for a secondhand Datahand Pro II)

[Matt Sergeant]

On Wed, 21 Oct 2009 16:32:15 +0100, James Laver wrote:
> On Wed, Oct 21, 2009 at 4:02 PM, Matt Sergeant
>  wrote:
>> 
>> True-ish. If you occasionally glance at the keys it really screws you
>> over though :)
> 
> Well unless you're buying labels to stick on the keys you aren't going
> to be able to look down on Dvorak, and I'd hope that by now you've
> muscle-memorised QWERTY.

On my own keyboard, yes. The point was on someone else's keyboard.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: keyboards/RSI/switching costs (was Looking for a secondhand Datahand Pro II)

[Matt Sergeant]

On Wed, 21 Oct 2009 10:32:46 -0400, jesse wrote:
>> Problem then comes with people who need to help you on your computer. I 
>> often help a tester here who has a "Natural" split keyboard, and find 
>> it tough, but doable (I used to use a natural years ago, the problem is 
>> using a Natural from a sideways position or standing position while at 
>> $co-worker's desk). I can't imagine any way of coping if he had a 
>> Dvorak layout.
> 
> When my friend Adam was switching over to Dvorak many years ago, he
> implemented two small shell scripts to toggle layout: "asdf" and "aoeu" 
> 
> It helped keep the problem Matt mentions in check.

True-ish. If you occasionally glance at the keys it really screws you 
over though :)

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: keyboards/RSI/switching costs (was Looking for a secondhand Datahand Pro II)

[Matt Sergeant]

On Wed, 21 Oct 2009 13:05:28 +0100, James Laver wrote:
> On Wed, Oct 21, 2009 at 12:31 PM, Chris Jack  wrote:
>> Before you switch keyboards, I think there is an important question 
>> about how often you are obliged to use a standard qwerty keyboard. I 
>> worked all over Europe for a bit using a large number of the 
>> European variations on qwerty (y and z switched for instance and 
>> punctuation in unusual places). I found the constant switching meant 
>> I was slower on all keyboards - but maybe it was worse because the 
>> keyboards were kind of the same. Maybe it's not such a problem if 
>> you switch between, say, qwerty and colemak.
> 
> A friend of mine in Canada tried it for a few weeks at work only ("I
> figured there was no chance of losing productivity at work") and used
> qwerty at home and seemed to do fine with switching. No subtle
> differences, it's a whole different mode of typing.

Problem then comes with people who need to help you on your computer. I 
often help a tester here who has a "Natural" split keyboard, and find 
it tough, but doable (I used to use a natural years ago, the problem is 
using a Natural from a sideways position or standing position while at 
$co-worker's desk). I can't imagine any way of coping if he had a 
Dvorak layout.

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Have at it

[Matt Sergeant]

On Mon, 26 Jan 2009 16:43:03 +, Denny wrote:
> As one is written by the EPO people, and the other by some random Perl
> geek, it might be safe to assume the EPO site knows best.  Unless you
> have information to the contrary.
> 
> If we're going to start contradicting definitive sources with things
> that people wrote on blogs, and classing them as equally valid, then the
> Internet will provide us with enough material to implode every brain on
> the planet.  :)

Indeed. Though personally I'd be really happy with the blog's 
conclusions if it were true. :-/

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

Re: Have at it

[Matt Sergeant]

On Sat, 24 Jan 2009 22:38:40 +, Simon Cozens wrote:
> 
http://www.dev411.com/blog/2009/01/14/perl-5-for-the-future-the-enlightened-perl-organization

Umm, this:

  << The goal is to modernize Perl 5 and make it competitive with new 
developments in programming languages, given that it's unknown when 
Christmas (the delivery date for Perl 6) will arrive >>

Doesn't jive with the goals listed on: http://www.enlightenedperl.org/ 
which appear to have nothing whatsoever to do with modernizing perl 5, 
but merely about promoting the values that perl 5 currently provides, 
and simplifying installation of software that works on the current perl 
5.

So which is it?

Matt.

__
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
__

[JOB] 2 developers

[Matt Sergeant]

Warning: This isn't a LONDON job. These jobs are in GLOUCESTER. It's just
that london.pm is the best UK place to post these things.

Here at MessageLabs in sunny Gloucester we need two more people to work
for us.

Neither of these jobs have anything to do with the web. Some of you
probably see this as a good thing ;-)

Position 1: General Development

We need someone to help our general development team out. The duties
involved here are writing perl modules, building RPMs, writing perl
applications etc. The work mainly involves understanding and working with
our email processing towers, developing modules that handle software
distribution to over 200 email servers situated around the world.

Position 2: System Architect

We need someone to help design and develop content filtering systems.
Previous experience with content filtering is a bonus. The candidate will
have pretty much complete control over how the system looks (though of
course it needs to fit into our current environment). The requirements are
pretty much written, but the design isn't.

The biggest point about this is that I'm not doing the hiring. So to
recommend someone I will need to know you, or know of you. So
participation in #london.pm is probably a bonus. I am involved in the
hiring process though, so be nice to me ;-)

Lack of specifics in the job postings is intentional.

-- 

<:->Get a smart net

Spare bed for me?

[Matt Sergeant]

Hi everyone!

I have just been informed by my boss that I'm expected to be in London
tomorrow at 9am to give a talk to our sales guys about anti-spam.
Unfortunately I live in Gloucester, where the trains are pretty crap from,
so I'm going to try and come down tonight. Amazingly this coincides with
the tech meet (annoying in one way as I'd have given a talk if I'd have
known), so it makes sense to come this evening.

However work won't stump up for accomodation. So does anyone live anywhere
near Oxford St that is going to the tech meet tonight that would be
willing to put me up for the night? Gratitude will be served in, erm,
perhaps a peek at the next version of AxKit?

-- 

<:->Get a smart net