Hackontest — 24h Open Source Coding Marathon

maemst writes "Can you code 24 hours non-stop? Hackontest is a new Google-sponsored 24-hour programming competition between different open source projects. Its goals are to enhance Free Software projects according to user needs and to make visible how enthusiastically open source software is being developed. During the current online selection process users and developers of open source software may submit feature requests and rate and comment them. On August 1st, 2008 the Hackontest jury will pick the three most promising teams. Each team will receive a free trip to Switzerland on September 24/25, 2008 to participate in the competition located in Zurich. Hacking 24 hours inside an etoy.CONTAINER, the teams and their virtually present communities will implement certain features based on the online ratings and jury selection. In the end, the Hackontest jury evaluates the code and awards the winners with a total of USD 8500. The jury is made up of 10 renowned open source contributors: Jeremy Alison (Samba), Jono Bacon (Ubuntu), Brian W. Fitzpatrick (Subversion), Martin F. Krafft (Debian), Alexander Limi (Plone), Federico Mena-Quintero (GNOME), Bram Moolenaar (vim), Bruce Perens (OSI founder), Lukas K. Smith (PHP) and Harald Welte (gpl-violations.org)."

Comments

Enhance?

[RAMMS EIN]

``Can you code 24 hours non-stop? ... to enhance Free Software projects''

I don't know about the rest of you, but, although I am sure I _could_ code non-stop for 24 hours, I am sure I won't be producing the best quality code if I do so. I think _enhancing_ any project is best done with clear thinking and sufficient breaks.

Re:Enhance?

[jellomizer]

the common argument ammong tech workers are the long hours... Yet we partisipate in competations to show it off. Mixed messages any one?

Re:Enhance?

[Devv]

I could probably hack for 24 hours non-stop but I wouldn't be able to work on a serious project for 24h non-stop
Obviously I am referring to the diffrence between a hack and a serious application built with continued developement in mind.

Re:Enhance?

[somersault]

Exactly, I tend to spend more time wondering which direction to best approach a problem from, trying to find an elegant solution that is succinct, but also flexible. I often spend a lot less time implementing and bugfixing code than I have planning it out.. there is more than one way to do it, as them perl people say - often the challenge is just deciding what is the 'best' way to do it for the application you are writing. I'd give examples but I'd just end up ranting for 2 paragraphs, as usual :p

Re:Enhance?

[flewp]

I don't know how it is for programmers, but as an artist I've stayed up for more than 24 hours, spending the majority of it on art (be it work, school, etc).

Once I started feeling tired, the quality of work suffered dramatically. No longer was I able to "go by feel" but had to actually think about the smallest detail, and usually it was for the worse.

Re:Enhance?

[SanityInAnarchy]

I'm not sure that's quite what GP meant.

I know that I can code, break, and code for some 18-20 hours, with very long breaks -- some to think about the project, and plan it out, and some to get my mind off the project entirely. But by then, the quality really does suffer, no matter how carefully I plan -- lack of sleep eventually makes me completely ineffective at anything, including coding.

I could probably do it with polyphasic sleep, but I'm not sure I have enough time to get on a polyphasic schedule before the contest.

Re:Enhance?

[Bruce Perens]

Nobody said the judges were being up for 24 hours. In fact, I made sure they knew I wasn't volunteering for sleep deprivation. And I just found out about this strange cyber-morturary container they propose to hold the contest in. Now, we know these things don't always get delivered, but if it does, I want a picture!

Bruce

Re:Enhance?

[modir]

The website states clearly that you have to be in a group of 3 people. So it depends on how you manage the team. Everyone has only to work for 8 or 9 hours if you plan it like this. Then as well from the website: "However, the Hackontest developers may connect to their outside community through chat, SVN, wikis etc. thus enlarging their team size virtually in a unlimited scale." In other words those in the container could only be the team leaders/project managers and those outside program.

Re:Enhance?

[Have Brain Will Rent]

Mod parent up - this is a ridiculous way to get new code into software. Sheesh, no wonder half the bug/security fixes I see coming through Ubuntu updates are buffer overflow vulnerabilities.

Re:Enhance?

[somersault]

Yeah sleep is good. I once had one of those moments where you're stuck on a problem, go to bed, and basically as soon as you wake up you've figured out the solution. If I have everything all planned out beforehand, then I think I'd be able to go for 24 hours (though in that time you could probably code a whole OS :P ), it seems kind of pointless though, the quality is bound to suffer by the end as you say.. and if the focus is on the actual coding without stopping to test everything thoroughly, then there will probably be a few nasty bugs too.

24 hours a day?

[moderators_are_w*nke]

Doesn't sound like a sustainable pace.

The question is...

[Devv]

Does the container have a toilet?

Re:The question is...

[Eevee1]

They'll be flushed with success.

Re:The question is...

[WK2]

Sure does. It's a hybrid toilet/deskchair.

Re:The question is...

[DeadDecoy]

They should, otherwise they'll suffer from buffer overflow errors.

int TOO_MANY_BURITTOS = 10;
byte poop[TOO_MANY_BURITTOS];

void codeToMuch(byte *poop){
    char *toilette;
    //toilette = (char*) malloc (TOO_MANY_BURITTOS * sizeof(byte));
    memmove(toilette, poop, TOO_MANY_BURITTOS * sizeof(byte));
}

There is no question...

[Yetihehe]

It WILL blend your brain.

Re:lol wat

[modir]

The link to the etoy.CONTAINER is not the best. The link goes to another project by etoy (which is based on the container idea as well). For more information about the containers you better visit this:
http://www.etoy.com/projects/etoy-tanks/

Publicity stunt

[mattMad]

To me this sounds like a publicity stunt with little useful output for the projects. I prefer the concept of the Google Summer of Code (even though many of the projects funded there seem to fail), because it focuses on a longer-term development and possibly recruits new talent to the projects.

Re:Publicity stunt

[Idiomatick]

Publicity stunt/extreme/nerdy is good in this case bringing hackers together for OSS to show off how hardcore they are. I doubt it has much to do with the code getting done in those 24hrs. Its an event.

Hopefully...

[John Hasler]

...none of the resulting code will make it into production. Marathon coding sessions produce only crap.

Creepy...ch

[fineghal]

Anyone else think the etoy.Container/Mission_Eternity link is creepy? Quote: "Mission Eternity is an information technology-driven cult of the dead." ... 'nuff said.

Re:Creepy...ch

[IamTheRealMike]

What I want to know is what text-to-speech engine was used in the video. Damn, that is the best one I heard yet.

Re:Creepy...ch

[Unoti]

Here is the speech synthesis you are looking for, and you can try it out right there on their web page. I love the software so much, and I've actually grown quite attached to her voice.

As soon as I read your comment, I thought I knew the answer even before hearing the voice. I then listened to the video, and sure enough, it was the voice I thought it would be.

I used to spend a fortune on books on CD, at $50 usd per book it became an expensive habit as I drove back and forth to work.

Then I started pirating books off a torrent site, then burning my own mp3's using speech synthesis software. I scoured the internets to find the best damn speech synthesis in the world, and I found the voice they used in that movie. I wrote software to take the text and semi intelligently split it into chapters, and make guesses about dividing the chapters down into 15 minute or so chunks for easy mp3 navigation. It has some features that let me preview what's going to get recorded so I can make sure it's naming the mp3 filesnames with the right chapter numbers. I've been steadily improving the software for a couple years. For text to speech, it just uses the Microsoft Speech API and I purchased that voice (Heather I think her name was) and it works amazingly.

The software isn't free, but it's so much better than anything else I've ever heard. It paid for itself after burning a few books to mp3. The voice is actually really expensive, I paid something like $150 2 years ago. However, at that time you could get a one month free trial, which is enough time to do a lot of playing and book recording.

Acapela offers several voices, but by far hers is the best. When you listen to her for hours, you'll hear traces of what I think is a castilian spanish accent. Acapela is based in Europe, and they do voices in many languages.

Anyway, their stuff is way better than anything else I've ever heard for general purpose synthesis. If there's anything better, I need to hear about it immediately, especially if it's open source.

Harder to cheat.

[SanityInAnarchy]

Put everyone in a box. That at least proves that they wrote it themselves.

What? Snow Crash?

[HiggsBison]

And I just found out about this strange cyber-morturary container they propose to hold the contest in.

Coffin hotels a la Snow Crash?

Re:What? Snow Crash?

[Bruce Perens]

Oops, it's not the cyber-mortuary, it's another product of eToys, also in a container. Coffin offices, sort of, but then freight containers aren't so small. That's reassuring. When I read the mortuary part, I wondered what sort of plans the organizers had for us :-)

Re:What? Snow Crash?

[Metasquares]

They're probably going to give the winners jobs. Spending 24 hours straight coding in a coffin is good preparation :)

Missing the point

[Enderandrew]

It seems everyone is missing the point. I doubt projects will be greatly improved more so in one day through this contest than they would be as opposed to normal contributions in a day. And sleep deprivation may lead to sloppy code. And the contest isn't inherently fair is people have internet access. But none of that is the point.

The prize is small, but the real reward here is PR for the OSS community, and these projects. OSS projects thrive on high visibility.

Re:Missing the point

[maemst]

Enderandrew, thanks for your comment - it's exactly as you write: We don't believe there will be huge code improvements within 24 hours - that's what the regular release processes are for. However, we believe FLOSS development needs more attention by the public (=everyone who doesn't read /.) Thus we'll try to get mass media (TV, news papers etc) to the event and document the Hackontest session and thus inform mortal software users that there are alternatives for their MS et al. products.

As second effect we expect to motivate FLOSS users to file their feature requests on the platform so developers get to know what people want. It's sort of addressing end user wishes with lot of demand (And yes, this concept is copied from Dell Ideastorm and Ubuntu Brainstorm - but now made available for all open source projects) Of course we are aware that there are roadmaps with future functionality enhancements. In this case it's even easier for developers to participate in the competition: Just file the features you've planned to implement anyway and maybe win a trip to Switzerland and some pocket money.

A last comment by the (unpaid) organizers: The first Hackontest is a trial - if many people participate in the selection process, file feature requests and rate them, if many skilled developers agree to implement features and show up at the event and if FLOSS communities support Hackontest overall (e.g. such as phpMyAdmin), then it's a success (and will be hopefully repeated next year). If not, we're sorry for the distortion we've created and won't bother you again with silly ideas ;)

Thanks for helping to make more marketing for great FLOSS products and participating on www.hackontest.org!

Re:lol wat

[Metasquares]

That page also has a poor explanation. My guess is that the company isn't all that great at clearly explaining what their products do.

Re:Vim

[SL Baur]

They asked me, but I turned them down. Better read between the words on that one.

(If) they (had) asked me, but I (would have) turned them down (because I think the whole idea is silly).

There, fixed that for myself.

Yeah, it's unfair that vim (which is a really bad version of vi, give me nvi any day) got representation and we didn't.

Re:lol wat

[modir]

Well, it is an art group. They talk like artists :)

I was in those containers. It is not that special. In one container are work spaces and in the other (the one they put on top) are some beds.

Re:Free as in Freeloaders

[Lally Singh]

I wouldn't be so hard on them -- most of them are misguided by idealism over the pragmatic realities of life. However I'm inclined to agree with your points.

I just wish you hadn't posted anonymously, so I filter up your posts!

Reminds me of MacHack

[Lally Singh]

MacHack was a 3 day/24hr conference with tech presentations. During the event, people put together hacks to compete for the covetous A-Trap award, given at the end. Some great hacks over the years.

I was at MacHack 19, good times. There never was a (real) 20.