Slashdot Mirror

Check out Capybara http://github.com/jnicklas/capybara complemented with the selenium webdriver. You can locally or remotely test any website in any flavor of web browser (great for that compatibility testing). Also the Cucumber team is doing a lot to make it work with Java, .Net, and a myriad of other languages.

and do some reading on "inversion of control"

Yeah, I forgot to mention that one. Inversion of Control is among the more useful concepts to come out of the software development community in recent years and it goes hand-in-hand with TDD and with modular programming and loose coupling with respect to software development in general. Personally, I use and like the Structure Map IoC framework (I especially like the assembly scanning and convention definition features), but Castle Windsor is also popular and even the Microsoft entry (late as usual) into this field, Unity, is fairly decent and getting better. The only caution I would give about using IoC is to take time to properly understand it before using it (it tends to be a bit of a mind-bender or an inversion, if you'll pardon the pun, the first time through) and learn to recognize where the boundaries are in your libraries and where IoC can help you achieve the goals of modularity and loose coupling.

Or another way to make it more fun - Godlike!

http://github.com/gunn/godlike

Plays a sound when you acheive 100% test passing rate. The default sound is the Unreal Tournament voice: GODLIKE, but you have as other sound options: headshot, killingspree, perfect, supreme_victory, ultrakill, flag_capture, and frag.

http://exogen.github.com/nose-achievements/

*
Instant Feedback
A suite of at least 50 tests takes less than a second to run.
*
Coffee Break
The suite takes between 5 and 15 minutes to run.
*
Take a Walk
The suite takes between 15 and 60 minutes to run.
*
Take a Nap
The suite takes between 1 and 5 hours to run.
*
Take a Vacation
The suite takes at least 3 days to run.
*

Coming soon!
Anticipation
One test in a suite of at least 10 is slow, and all the rest pass.

Time

*
Night Shift
Make a failing suite pass between midnight and 5am.
*
Punctuality
Make a failing suite pass at 9am (give or take a minute).

Failure

*
Complete Failure
All tests in a suite of 50 to 999 fail.
*
Epic Fail
All tests in a suite of at least 1,000 fail.
*
Minor Letdown
All tests in a suite of 10 to 99 pass...except the last.
*
Major Letdown
All tests in a suite of at least 100 pass...except the last.
*
Happy Ending
All tests in a suite of at least 10 fail...except the last.
*

Coming soon!
Heisenbug
Make a passing suite fail without changing anything.

Errors

*
Sausage Fingers
Cause at least 2 distinct syntax errors in a single run.
*
To Understand Recursion
Exceed the maximum recursion depth.
*

Coming soon!
To Err is Human
Cause all tests in a suite of at least 50 to fail with a single error.

Size

*
My God, It’s Full of Dots
The suite has at least 2,001 passing tests.

Frequency

*

Coming soon!
OCD
Run the suite at least 5 times in the spa

Yes, agreed, a combination is good (SQL + NoSQL + filesystem).

There is no one-size-fits-all scenario, here.

However, there is utility in a NoSQL database over a raw filesystem. One feature is indexed search. Another is versioning. Another is the fact that it is extremely multiuser (proper record locking, even if there are multiple writes to the same record). Also, many NoSQL databases (especially MongoDB) have built-in replication, sharding, Map-Reduce, and horizontal scaling.

MongoDB's GridFS (especially with FUSE support) marries many of these features together. MongoDB does have some SQL DB features (such as indexing/searching and transactions) but not others.

Check out the whole stack here:
    http://www.mongodb.org/
    http://www.mongodb.org/display/DOCS/GridFS
    http://github.com/mikejs/gridfs-fuse

Given the history of web development, dynamic languages became widely used mainly out of ignorance, and have remained widely used due to continuing ignorance. There's no technical argument in favor of dynamic languages. They're just used because their users and proponents often don't even know about how much better and easier static languages make the development of both small and large applications.

While I agree with much of your comment, I believe there are reasons for the continued use of scripting languages in web dev, other than ignorance or stupidity.

The web moves fast, and scripting languages, with their lack of compile-run cycle allow quicker development. Things change enough that this is often worth the corresponding drop in execution speed.

Secondly, the widely-used statically-typed languages (Java, C#) are about as non-DRY as you can get, Java in particular. While static typing does not necessitate needless boilerplate, the languages that remove it are only now starting to gain acceptance, and haven't gotten a firm hold on the market (or strong communities) yet.

Thirdly, shared web hosts (where the majority of small businesses and personal sites are hosted) often don't support much other than static files and PHP, possibly ASP (.NET or otherwise, I haven't checked in a while). I've recently switched to doing a lot more static sites generated by Jekyll, simply because GitHub will host it for me for free (and only having to push my git commits to deploy is nice, too). Hosting matters.

Finally, I find that community is far more important than the language itself, and the innovating part of the web community is a fickle beast, but tends to stay in the realm of dynamically-typed languages. It takes a few years to get something roughly as nice as Rails (well... other people like it) or Django, and years are decades in our field. Things like HAML have been ported to .NET and Java, but the development's done in Ruby.

Given the history of web development, dynamic languages became widely used mainly out of ignorance, and have remained widely used due to continuing ignorance. There's no technical argument in favor of dynamic languages. They're just used because their users and proponents often don't even know about how much better and easier static languages make the development of both small and large applications.

While I agree with much of your comment, I believe there are reasons for the continued use of scripting languages in web dev, other than ignorance or stupidity.

The web moves fast, and scripting languages, with their lack of compile-run cycle allow quicker development. Things change enough that this is often worth the corresponding drop in execution speed.

Secondly, the widely-used statically-typed languages (Java, C#) are about as non-DRY as you can get, Java in particular. While static typing does not necessitate needless boilerplate, the languages that remove it are only now starting to gain acceptance, and haven't gotten a firm hold on the market (or strong communities) yet.

Thirdly, shared web hosts (where the majority of small businesses and personal sites are hosted) often don't support much other than static files and PHP, possibly ASP (.NET or otherwise, I haven't checked in a while). I've recently switched to doing a lot more static sites generated by Jekyll, simply because GitHub will host it for me for free (and only having to push my git commits to deploy is nice, too). Hosting matters.

Finally, I find that community is far more important than the language itself, and the innovating part of the web community is a fickle beast, but tends to stay in the realm of dynamically-typed languages. It takes a few years to get something roughly as nice as Rails (well... other people like it) or Django, and years are decades in our field. Things like HAML have been ported to .NET and Java, but the development's done in Ruby.

This is exactly why Facebook optimized PHP (with hiphop: http://wiki.github.com/facebook/hiphop-php/) last.

It really was the least import part when optmizing.

After the release of all of the publicly available facebook data last week there's a real question of who owns your data, and if it's public does that mean it should be publicly accessible via open APIs for anyone to grok? an immature but developing example is things like phasebook http://github.com/philcryer/phasebook

This is Frash, in fact the link that should have been included in the story should have been to the Frash github (this IS slashdot after all not gizmodo) http://github.com/comex/frash

Also until recently Frash only ran on iPad, like you state, but now it is getting better on iPod Touches and iPhones of a couple generations as well, so that's new and worthwhile info.

You can watch for the code the repository in github. It's empty at the moment. http://github.com/Burton/Analysis-of-Competing-Hypotheses

Here you go: http://wiki.github.com/behlendorf/zfs/

I understand why, but there are a ton of people out there that think OSS is OSS. You wonder why corporations are weary of OSS it's because of this. I really hope this project goes somewhere or Debian's kFreeBSD project works as well as I'm hoping.

Reminds me of this joke:
I was walking across a bridge one day, and I saw a software developer standing on the edge, about to jump off. I immediately ran over and said "Stop! Don't do it!"

"Why shouldn't I?" he said.

I said, "Well, there's so much to live for!"

"Like what?"

"Well ... do you develop Closed Source or Open?"

"Open."

"Me too! Are you BSD or GPL?"

"GPL."

"Me too! Are you GPL v2 or GPL v3?"

"GPL v3!"

To which I said, "Die, heretic scum!" and pushed him off.

I've got to second that. It's not just that most of the projects which use it don't need it, it's that it's a kludge on top of a kludge on top of a kludge that makes IBM's zOS look good.

It's not just that DBAs perform tasks which would otherwise be done by developers or sysadmins. It's that Oracle seems to almost actively encourage the DBA as a profession. A trivial example is the autoincrement column -- even sqlite has one built in, but no, on Oracle, you have to create a sequence first, then that primary key column, then a trigger that inserts the next value from the sequence.

And it's not just the database. The kind of products Oracle tends to acquire seem to be structured to encourage huge apps, any one part of which can be managed by an entire department. I'm just finishing up a project in ADF for the summer, and I've had to use seven distinct languages (a conservative estimate) to do it, woven into a stack as many layers deep.

And you'll have to implement hundreds of random hacks to bring it up to the level of what you get for free with other, competing products, and at the end of it all... My application crashed WebLogic.

No, let me say that again.

My application made WebLogic segfault.

WTF? The entire fucking stack is in a JVM, except the Oracle database itself. How do you segfault that? Unless they're running native code somewhere to optimize it, which would be really sad, considering it takes longer for my app to spin up (once WebLogic is running), and takes at least four to five times as much RAM, as a comparable Ruby on Rails app. But comparisons aside, I managed to segfault WebLogic with application code. I wasn't doing anything that tricky, I just slightly misconfigured ADF...

So not only are they huge, bloated, make-work products, in the sense that their entire purpose seems to be to create little niches of niches of jobs supporting this Enterprise-Level Shit, but it doesn't fucking work.

Yet they sell like hotcakes, because they're enterprisey, whatever that means.

I don't think they're good at marketing in the traditional sense. At this point, I really, honestly, truly suspect blackjack and hookers are responsible for these kinds of business decisions, because I can't see a rational agent wasting that much time and money doing it so horribly wrong.

You are almost certainly thinking of "Journey to Wild Divine"...

A bit on the woo-tastic side(Deepak Chopra is usually a bad sign); but I certainly don't know of any other myst-like biofeedback games.

Someone has also hacked together an OSS driver for the biofeedback peripheral, in case you prefer stats to swamis.

Rakudo Star includes a PDF of the Perl 6 introduction book; the print version should come out late next month. Moritz Lenz's Perl 5 to 6 article series is always informative and useful. The official Perl 6 site's documentation page links to current and accurate documentation.

Normally i'm doing dvd some progressive, some interlaced/telecine'ed content, some of it is interlaced, most isn't. I was under the impression that -deinterlace didn't do anything if the content wasn't interlaced. I'll look into the -top option. The full python code can be found at, http://github.com/cynyr/ps3enc It is a huge mess i know.

Go has the same problems. They try to make it 'simpler' but along the way they actually make it more complex.

For example, try-catch-finally idiom is an easy and standard way to deal with exceptions. But no, they had to invent their own half-assed implementation just to be 'minimal'.

Also, they insist on using fucking _return_ _codes_ to indicate errors. WTF? It only makes code more complex because of tons of stupid 'if error' statements.

Personally, I like Rust's ( http://wiki.github.com/graydon/rust/project-faq ) design more. At least, it has some new features.

I posted an article a while back about a DHTML engine I put on GitHub. I included an example game called Bombada that's also on GitHub. Note: the engine isn't "HTML5" per se (which is becoming more of a buzzword than makes sense) and I've moved on to a canvas engine (which will someday also be open source).

Even better, there was recently a game development contest on Boing Boing which saw 9 pretty cool entires. Ours was called Onslaught! and was written in JavaScript using canvas (though it does fall back to flash for audio).

I've got some substantial experience writing games in JavaScript and HTML5. To me, the biggest hurdle right now is audio. Somebody mentioned the inability to go fullscreen, and while I've seen that handled by the video tag, to me it's not as big a problem as the audio tag being basically unusable for gaming purposes.

I posted an article a while back about a DHTML engine I put on GitHub. I included an example game called Bombada that's also on GitHub. Note: the engine isn't "HTML5" per se (which is becoming more of a buzzword than makes sense) and I've moved on to a canvas engine (which will someday also be open source).

Even better, there was recently a game development contest on Boing Boing which saw 9 pretty cool entires. Ours was called Onslaught! and was written in JavaScript using canvas (though it does fall back to flash for audio).

I've got some substantial experience writing games in JavaScript and HTML5. To me, the biggest hurdle right now is audio. Somebody mentioned the inability to go fullscreen, and while I've seen that handled by the video tag, to me it's not as big a problem as the audio tag being basically unusable for gaming purposes.

Find project you like or use and start contributing. Or ask them if they need any help.

Most of the big ones do have "help us here" pages, such as KDE:
http://techbase.kde.org/Contribute

And another KDE page for those just starting out:
http://techbase.kde.org/Contribute/Junior_Jobs

So either the OP needs those links, or he is looking for smaller projects to help with. Here, let me suggest some small-project tools that I use that could use the help:
Anki, flash card application: http://ichi2.net/anki/index.html
Zim, desktop wiki: http://zim-wiki.org/
Gmail Conversation View for Thunderbird: http://github.com/protz/GMail-Conversation-View/issues
Vimperator/Muttator: http://vimperator.org/
Redshift, change screen colour per time of day: http://jonls.dk/redshift/

While it won't help for onroad routing, Pincaster might help you to store 2d geographic data and to easily display it on a map : http://github.com/jedisct1/Pincaster/blob/master/README.markdown

Apple says Flash is welcome... as long as it is implemented in HTML5. And so that is what people did. Everyone except for Adobe themselves.

Really goes to show that Adobe does not stand behind their platform. They are only in it to milk their existing codebase.

tl;dr, but: No browser today does printing really well. If you want to accurately get out of a web browser what you see in the web browser, your best bet is to effectively generate documents that are native to paper, like PDF. Tell your devs to look at PDFKit. It uses WebKit to generate PDFs on the backend. Built in Ruby and includes Rails integration and a Rack Middleware. Works as advertised.

you might want to give serious consideration to creating an application that uses the web engine itself, directly, rather than relies on the broken-ness of the web browser "applications".

two possibilities for you where you can start: http://wiki.python.org/moin/WebBrowserProgramming if you prefer python, or if you truly prefer ruby, try http://github.com/ppibburr/rbwebkitgtk which is ruby bindings to webkit *including* direct access to the 2,000+ functions of the DOM model.

in this way, you can set up an application (ruby-gtk+ app) where you will be able to have your *own* menubar at the top - not the stupid web browser's one; on it, you would have a "Print" option which went and accessed the DOM model's "innerHTML", and you passed it to a *decent* printing system under *your* control, using whatever printing methodology is accessible to you via ruby, running on the user's machine.

or python, if you prefer.

this type of methodology seems to have been entirely forgotten about, as people scramble for conversion of complex apps to "entirely web-based" because they were convinced that "web 2.0 is Good and is God", and then run head-long into exactly the difficulties you're encountering, by which time it's too late.

advice: bypass the web browser as a de-facto application, and create your own application still using the incredible power of web browser *engines*.

there is also prawn for pdf generation with ruby.

If you don't want to learn Prawn and are more comfortable with HTML, use something like PDFKit, which is a wrapper around wkhtmltopdf, which basically just packages WebKit for PDF printing. You'll be able to use CSS3 and other niceties and it'll be consistent, since it's creating a pdf. Perhaps you can do better things with Prawn, but I guarantee that writing moderately complex reports will be easier with PDFKit then with Prawn.

Reaching a bit, that means that a group of like minded people could trade apps around, in source form, which each member could build in their individual development environments and install on their phones.

For instance, a free proxy app that any developer can run and install (Cydia of course has much nicer proxying applications that people can buy).

I admit I haven't tried this out but there is a native port of zfs to linux:

http://wiki.github.com/behlendorf/zfs/

There is an "issues" tab with 20 open issues but they aren't loading for me.

It's been forked - latest change was 4 days ago

http://github.com/gustavosbarreto/antico

I know that it isn't always the smoothest, but Facebook is PHP, and is pretty fricken huge.

Sort of. According to what an employee told me last week, lots of the core services are written in other languages, and then PHP is the glue that queries those services and builds the page itself. They then have found that that really isn't fast enough for what they want, so they've written their own compiler (Hip Hop), which compiles PHP down to C++ which can then be compiled by G++.

So yeah, facebook uses it. But not exactly like you'd expect.

No. You're wrong. It doesn't use just Javascript to determine your location. The applications have access to wireless services on each OS and sends those to google. You can even craft a string to send to google.

Look at how PreyProject uses it:
http://github.com/tomas/prey-modules/blob/master/geo/core/functions

Or run through a proxy server in another country and try geolocation as described above. It'll still find you no matter your external IP.

CANVAS is now viable for use, even with IE.

Here's a benchmark of SWF, HTML 4 DOM, SVG DOM, and HTML5 <canvas>. How fast does IE render <canvas> compared to Flash Player? Or did you mean Chrome Frame?

Also,just as mini me mentioned in reply to you, Smokescreen can just convert it directly from the SWF to browser-based implementation.

Then you're still using Flash format, just with a different player. Some HTML5 partisans want to get away from Adobe software entirely.

There is also the Gordon library.

(Google gordon library) You don't mean the WPI George C. Gordon Library or the Gordon Community Library and Museum. (Google gordon library javascript) Nor do you mean the E. Allen Gordon Library. Oh, you meant this Gordon, which does much the same thing as Smokescreen.

Take two if you find a C file in a #include. http://github.com/blinry/gish/blob/master/main.c

I've found that's the best way to merge SQLite's core (a multi-megabyte .c file) with its (commercial) encryption core without having to cat the two together each time a new one is released. Especially on platforms like windows which don't have cat.

Actually, it's possible that they're doing something similar to the SQLite amalgamation, i.e. trying to pack as much of it into a single file as they can to improve the global optimisations. Of course, they might simply be on crack.

Take two if you find a C file in a #include.
http://github.com/blinry/gish/blob/master/main.c

See http://github.com/RHoK/

I don't know what the other teams did, but the team I was on intended to release under this license: http://www.opensource.org/licenses/nasa1.3.php

is an idiomatic expression the purpose of which is to divert the audience from the truth or an item of significance. For example, in mystery fiction, an innocent party may be purposefully cast as highly suspicious through emphasis or descriptive techniques; attention is drawn away from the true guilty party.
---
How is this any different than my bank forcing me to get an 'authorization code' via Text every time I login with a computer that doesn't have their cookie set?

The ninth highest search on Google is "delete facebook account"

Looks like the house of cards is starting to crumble. I know it's stupid, but maybe if they kept it simple like back in the day.... (Although I love the API for batch uploading photos)

The iPhone is already more than capable of running Flash without violating any of Apple's rules, as seen with the Gordon project. If a couple of people can get a basic Flash interpreter working in their spare time, surely Adobe can put some man-hours towards a full implementation.

There is no question that Apple is making it difficult for Adobe, but the truth is that Adobe is not serious about Flash. They want a free ride with their existing products, but have no intention of doing whatever it takes to bring Flash to devices that require a little extra work. It is easier to go crying to the media and make another company look bad for Adobe's own shortcomings.

I do not like Apple's restrictions as much as the next guy, but Adobe needs to get over it. Either bring Flash to the iPhone through the allowable channels, or make Flash so compelling on other devices that nobody will consider purchasing an iPhone in the future.

It claims to support SWF1 and a lot of SWF2. Right now I believe we're on SWF9, so there's a long way to go, but it does show that the approach works.

One is compiled then executed in a VM; the other is already compiled and executed in a VM. In the optimal version of each, Javascript will be slower.

Presumably you have the benchmarks to validate your claims? JavaScript was beating Flash two years ago:

http://fupeg.blogspot.com/2008/09/javascript-faster-than-flash.html

Plenty of people are actively migrating away from Flash. Here's an example:

http://github.com/blog/621-bye-bye-flash-network-graph-is-now-canvas

The problem for Flash is that it's going up against browsers with highly optimised JavaScript engines with soon to be hardware accelerated render engines (see Firefox's progress for example). It's all looking a bit grim for Flash.

Actually, the official page does mention it, under "Unofficial implementations":

Vim (crossplatform) - Sparkup, Zen Coding for Vim

Check the TUTORIAL file in the repo to either turn on by filetype or globally

No, flash plugin doesn't work perfectly fine on other macs. It is a resource hog and incredibly buggy. It's the only reason Safari has crashed in the past year or two. Since I installed click to flash, I haven't had a crash. Well, that's entirely not true. Once in a while I let flash run on certain sites. When I do that, there's always the possibility of a crash. On top of that, flash on a mac is as bad as the linux implementation. It's slow. For some non video sites, certain flash animations can use more than one core. Usually those same sites, do not use that much cpu power with the windows plug in. Flash on a mac is really really awful.

Well obviously you are right that for some people cash only is the best way for them to control spending. My point though is that anyone with a bit of self control is benefited by using a credit card for all their transactions.

The number one rule is to not actually buy things on credit, and pay off all of your debt at the end of every billing cycle. If you can commit to doing that then credit cards give you a lot of power for tracking your spending. A lot of people don't track their expenditures because it takes a lot of work, you have to at some point sit down and check the balances of your money pools, go to the back to withdraw money, etc.

With a credit card, or even a debit card it becomes a very simple matter to download your statement every month and import it into Quicken or Microsoft Money or whatever. When you make the barrier to entry easier more people will track their finances.

No matter what method you are using though, cash or debit or credit, the important thing is to have a system and know where your money is going. People who don't budget and spend without thinking or tracking are going to get in trouble no matter what method of spending they choose.

My system it to pay all my expenses with a credit card, and download the csv of my statement from my bank. I have a ~500 line perl script that parses it and write ledger entries for all my expenses. From there is is very simple to track my expenses vs my income, build a budget and track how closely I stick to it. I have seen a huge improvement in my personal understating of my finances and spending habits, but of course not every method is good for every person ymmv.

Actually, it's more like: INTERNET -> PORT22, since just about anything can be sent through an ssh tunnel. And the encryption makes most types of deep packet inspection impossible.

You missed his point which wasn't about the protocol, but the port being used. If you use port 22, it'll be blocked many places because they don't want to allow you to ssh. If you use port 443 it'll be allowed since https is "necessary", even if you're using 443 to carry your ssh traffic. What's sad is seeing other services move to 443 to be more accessible. Most usenet providers offer SSL encrypted NNTP on port 443 (despite having an RFC port specifically for nntps).

But it is much harder to block if they actually use legitimate looking packets for protocols that get out rather then just it's port. So people have encapsulated IP within real HTTP traffic. Better yet they'll use ICMP or even DNS to carry your traffic. I find the DNS one particularly amusing because it uses your nameserver to redirect the traffic even if the host isn't given any outside access.

An exception probably exists for the people who wrote a new PHP engine for facebook.

Are you referring to HipHop?

HipHop transforms your PHP source code into highly optimized C++ and then compiles it with g++ to build binary files.

No JIT going on here.

career experts say you have to take a strategic approach to your job search and application process. The best candidates are always taking steps to manage their careers...

I fully agree. If you sit passively and wait for your next raise, you may be waiting for a while... But if you are proactive, good things eventually happen to you. Contribute to an open-source project. Become the co-founder of a cool iPad app or whatever cool idea people are trading nowdays...

It doesn't pay off instantly, but a year or two later, your resume stands out from the crowd, and more importantly, you may not even need a resume anymore to get a great job!

The actual code is here: http://github.com/cantrell
The platform specific files are configuration files and icons and such.

Stability? Maturity?

Basically, this page has a large and scary list of stuff that's not done yet, but is in the basic language spec. It's also pretty clear they're nowhere near 1.0.

I admit, that's much farther along than any Perl6 was last I checked, but contrast to, say, JRuby, which is in version 1.4, has access to tons of Java libraries (Does Rakudo connect me to Perl5 libraries from CPAN?), and is, in general, actually production-ready.

Here you go