Projects that should have their own websites
There are plenty of interesting and valuable projects on CPAN that do not have their own websites. I know, I know, CPAN gives us a default website ("with a reasonable design") and all we need to do is write the code and POD. Yes, that's great.
But wouldn't you rather advertise yourself, your product and Perl along the way? Then your project should have its own website. Yes, it should. No, I'm right!
A website represents you and your project. If it's beautiful, you're beautiful (even if you're ugly). If it's approachable, your project is approachable. If it gives a warm feeling to the user, so does your project and so will your project be for the user's boss - at least at first. Which means that if you're certain your code is great and approachable ("once given a shot"), you should make sure it gets a fair shot, by making it appealing.
Projects that should have their own websites:
PPI
WWW::Mechanize
DBIx::Class
Module::Starter (I actually wrote one, but never got it up and running)
Dist::Zilla
Module::Build
Module::Install
AnyEvent
CPANPLUS
App::cpanminus
Test::Simple
ORLite
Projects that already have websites, but should be revamped:
Dancer (working on it)
Padre
ack
POE
PSGI
CGI::App
Honorable mentions:
Mojo
Catalyst
Template Toolkit
KiokuDB
Moose
Specifically KiokuDB and Moose have pretty sweet designs. However, they're both simple pages, and they deserve an entire website.
The best way, IMHO, to get a beautiful website, is to hire a professional web designer. Of course, we're all hardcore DIYs and we like doing things our own way and we can handle some {X,}HTML, but unless we're professional in this field, we're probably better off hiring someone.
I might know how to hammer in a nail, but I won't be building my own house, unless I know what I'm doing (and I can assure you I don't). So, in short, hire a professional to do a professional job.
Gots to do the bloggingz
See also
Marketing & CPAN – mutually exclusive? and
Marketing & CPAN - follow up and an example of changes by Oliver Charles.
Thank you
http://code.google.com/p/www-mechanize/
That isn't a website. It's a small wiki. That is like saying a Github page is a website. It isn't.
I recently made a minimal website for Hailo which just uses the GitHub CSS + GitHub pages.
I've set out to create websites for something or other several times and what usually blocks me is trying to make it look good.
I used this design for osm.is but usually I'll either just give up or use no design at all.
I've tried to search around for a library of free (as in freedom) nice looking design templates but I haven't really found any. Is there something obvious I'm missing or are these in short supply?
> Is there something [...] I'm missing [...] ?
Yes, there is. Hiring a professional. As hard as it is for us to admit (myself included!), we should hire professional web designers to do professional web design.
At least for stuff we want to show other people and be proud of. Anything that relates to our jobs (personal or corporate) and anything that can degrade whatever technology we use if not done right.
If you mean that people who offer freely code, docs, tutorials, articles... to the community, should also pay to have a professional looking website, then that doesn't sound quite reasonable to me.
If you're talking about getting some of the users of those tools to give money or designer time to create or improve their web presence, then it makes sense.
How about an "Adopt a Module" program?
I do mean that people who freely offer code, docs, tutorials and articles to the community should also pay to have a professional looking website. I do mean that. Let me explain why:
Your "free code" isn't a random assortment of statements (at least I hope it isn't). It's a product you create, whether for free or for pay. The value of this product is not just monetary. It represents you. It's your resume, it's your portfolio, it's your interest, communal contribution and future job assurance.
We should burst out of the "it's free, leave me alone" bubble. My job depends on Perl programs and modules being available. That way I have a competitive edge and I can provide to a company that hires me a reliable service. It's perfectly logical for me to contribute to a project (either time or money).
If it's a project of mine, it's my face, and I would want to promote that project as well.
Another interesting point is that people often make a very strong effort at making things pretty, and put a lot of hours and effort into it. Why? You could be saving up the money of that spare time you're wasting and pay someone for it. S/he will most likely even do a better job than you.
And what category do you want to put strawberryperl.com in? Or does it even count? (It's the website for Perl::Dist::WiX, Perl::Dist::Strawberry, and Perl::Dist::Chocolate (although in this case, it's the output of running those modules that is featured, and not the modules themselves.)
I understand what you're saying but there's a significant leap from "I have this little project I work on in my free time" to "I'd like to spend lots of money finding / hiring / managing someone who'll make a pretty website design for said project".
Some of the projects you list like Catalyst have enough shared commercial interest to pull something like this off. But a lot of "1-hour after work" Perl projects don't.
Those could be significantly helped if
search.cpan.orgpresented a friendlier interface that could be customized on a per-project basis. Unfortunately there are non-technical hurdles blocking that path.As for the thing I was missing I've found it in Open Source Web Design which brian d foy pointed out to me.
The Catalyst and Mojolicious sites (including icons) were donated.
There is also BioPerl.
I agree and disagree.
I realize that you haven't talked about the URLs so this might sound like a moot discussion, but having pages on googlecode or gihub does have a benefit that the page (domain) will be less likely to expire once you lose your interest. I have some custom domains for my perl applications such as http://plagger.org and http://remediecode.org but i'm not really sure if I keep those domains for another 10 years.
Real world example: juno.py, the sinatra-ish minimalist python framework page is now 404 http://brianreily.com/project/juno/
Also, while i agree that the design on search.cpan.org is not too newbie friendly and could be improved, it depends what are the user targets of the individual projects. Tools such as Plack, AnyEvent or ack totally target programmers, not suits, so even if they should have its own website, the site should be more focused on documentations and community forums type of things rather than "professional" shiny designs.
Examples:
Unicorn the Rack HTTP server http://unicorn.bogomips.org/
EventMachine the Ruby event-loop library http://rubyeventmachine.com/
WSGI http://wsgi.org/wsgi/
They have their own websites, but it's just a set of documentations and wiki pages, and has no shiny designs, and they're successful.
My 2c.
I agree that some of these applications need their own web sites. It would be preferable to find someone who is enthusiastic about the project and is a good web designer to build the site but if you cannot find one then paying a professional is indeed a good idea.
I have no idea how much such a good designer would cost or how to manager her - I personally never worked with a web designer that you can also see on my web sites and I don't trust myself to be able to say which one would look good and which is not - but maybe this is something the TPF could finance. After all having a few very good looking web sites for various Perl projects would make a really good job for Perl advocacy. The question still remains who decides if the particular design is good or not?
I would also say that the domain names of these sites could be registered by the TPF and paid by the TPF if the author wants that. That would hopefully ensure continuity beyond the interest of the initial developer of the project.
StrawberryPerl.com (to me) is in the Very Honourable Mentions.
That is because it has a very simplistic - yet approachable - website. It's beautiful, it's slick, it's pretty. StrawberryPerl.com is an example I often give to people on how to take a serious product and present it in a very slick, clean, simple and compelling manner.
I <3 StrawberryPerl.com
Maybe it's just me but i can't take web frameworks serious that need to pay for web design.
Quite the contrary!
A web framework is the underlying of a website, it's the core functionality. It's not the design, and it's a shame you think of it that way.
A web framework that pays for web design, is one that didn't get any by donation and understands the frontend layer (the View, the design) is just as important and gets a professional to do it. They themselves put their time and effort into what they do - the code of the framework itself.
Your sentence should be "Maybe it's just me but I take web frameworks that pay for web design more seriously."
Not at all, you misunderstood me there, web development is as much frontend work as it is backend these days, the line gets even more blurry with HTML5.
If framework developers can't do both i don't think they should be building frameworks, simple as that.
I respectfully disagree.
One of the mistakes PHP made was to think the developer is also doing the design and the designer also does the development. So PHP was actually HTML with embedded coding.
Since then, MVC was developed and people realized templates are the way to go because they allow two different jobs: a designer and a coder. Now PHP had to revert, so these days PHP is embedded code in HTML that runs a template to render HTML written elsewhere, so it's on backwards. Kind of funny.
The lesson was though, that complexity grows, and when before the job was "developer" and you wrote code, graphics and web design, now they're different jobs.
Think of gaming development. Do you think the person who writes the engine designs the characters? Do you think that a person who can write a complex engine with various levels and stages, supporting several graphic cards should be out of job because he can't design a character worth a shit?
You skipped HTML::Mason’s nicely functional website.
Are you even following current web development trends?
There is a huge movement forming to promote JavaScript on the server side specifically to bring frontend and backend development together, and they have very good chances to succeed. (see node.js)
You might have had a point 10 years ago, but times have changed.
“Most people make the mistake of thinking design is what it looks like. People think it’s this veneer – that the designers are handed this box and told, “Make it look good!” That’s not what we think design is. It’s not just what it looks like and feels like. Design is how it works.” —Steve Jobs
I concur with Anonymous. (That you go and talk about PHP and MVC and all that tells me you completely missed his point.) A good web framework is one written by people who design sites (in the “how it works” sense) all day long. These people may not have great design skills themselves (in the “how it looks” sense) but they almost certainly have good access to designers who care and are willing to oblige them. And they also have a sense for the particular recurring requirements of producing well-designed sites, which is likely to affect their frameworks’ technical design (think of the many kinds of helpers in Rails). So while an ugly site for a web framework does not automatically mean it is bad, it does indicate that the web framework was built by someone who likely has a somewhat shaky sense of the needs in building well-designed sites.
Which by your own admission in a more recent post of yours is basically exactly the case for you.
I think we're talking about completely different things, and we're using the same word to describe two different things as well, so it's even worse.