BAWAHAHAHAHA! Hilarious!

Any of you good with software licencing? I have a program that I potentially want to test in the wild with some handpicked customers but I don't want it to be free/OS during the test phase. Once the results are in I want to change it to a free/OS license and force them to autochange their inplemented software to that licence as well.

I guess the best way would be to have it licenced under the free/OS license I want all along and have them sign a trade secret type of contract that they cannot share the software with anyone untill it's officially released or something?

I'm not decided on BSD or GPL type yet if I go with a BSD type I suppose I could just install a proprietary fork of the codebase I hold under BSD internally and use that for the testrun then later update their systems to a BSD licenced one (assuming they'd agree/update)..this path seems like a very ugly hack though

Not sure if I'm making clear what I'm trying to do here. I'm quite opposed to IP of all sorts so having to use such a strange construct is hurting me quite a bit (but then again I can't legally public domain software here so I have to resort to IP either way, free/OS licenses fall under IP ldo).
I think "sign a contract not to discolose software details"+some free/OS license would "feel best"

Sounds like you need them to sign a Non Disclosure Agreement (NDA) if they want to be part of the beta. I'm not sure how NDA on a GPL'd piece of software works. They kind of solve 2 different issues so I wouldn't expect the GPL to interfere (override, etc) the NDA. I'm pretty noob when it comes to licenses as well.

Ninja Edit: Looks like GPL does interfere with an NDA

http://www.gnu.org/licenses/gpl-faq....TheGPLAllowNDA

clowntable if you are the sole author of this software you can release the same thing under as many licenses as you like pretty much - no reason you can't NDA people to test binaries under a top-secret license, then later release the exact same thing as GPL.

The main purpose of an NDA can usually be summarized as "Don't talk about it!"

I'm not sure if I understood you correctly but if all you want is to not have anyone distribute or build upon your software during the testing phase, then I would suggest not doing it the way you described it in the quote above, but instead just allow as little as necessary during the testing period (with no foss license yet) and you can without any problem allow everything you want later.

But I'm not sure, so one question:
Why is it that you can't put it in the public domain if you want to? Or do you just mean not during the testing phase.

Just as _dave_ said you can release it under as many licenses at the same time as you want to. Basically the only thing you can't is give someone the software under a license that grants rights like the gpl and then later want to take the rights from that person away.

But you could still release it under a proprietary license since you own it.

If later you want to have it as free as possible I would suggest against the GPL, since it does come with restrictions but rather use the WFPL or a similar version without profanity.

clown,

i think others have said this but i'll answer as well and maybe clarify:

- release code to beta testers under $restrictive_license
- once ready, release to everyone under $foss_license
- you can't change the license agreement you already have with your beta testers, but it doesn't matter because they can at any time use your software under the (less restrictive) $foss_license

The concept of public domain does not exist in German law. It's really dumb and I'm not aware of any other country that has this except Austria.
You automatically get rights to stuff you create which you can't explicitly waive and they end 70 years after your death.

Guess I might have to consult a lawyer

re: the Javascript loop talk earlier:

http://blogs.oracle.com/greimer/resource/loop-test.html

Have any of you been following the recent buzz about kickstarter? Double Fine broke all records by raising over $1MILL in 24 hours. This is a big step for croudsourcing imo.

http://www.kickstarter.com/projects/...fine-adventure

Been used a lot for board game development for quite a bit now.

My printer is tilting me pretty badly, it blue screens windows after each page it prints, printing a 6 page document takes about 15 mins with 6 restarts lol.

Wow, how is that even possible? I know its windows but man, that happened to me there would be a fair chance I'd have to buy a new printer. And possibly a new keyboard and maybe screen too...

No idea lol, it's a joke. Reinstalling the drivers would probably help but I cba, they are big bloated files server from a server that lets you download at about 20kb/s. Takes a while to configure the software it to not spread like cancer over your computer as well with reward popups and alike.

godamn it I have to print 42 pages, ffs lol I never have to print anything usually

never work with children, animals, or physical hardware.

crosspost from sports betting, re: database design

dunno if this is the right place to ask, i'm trying to figure out how to design a database to keep a log of sports odds throughout the day. i plan to scrape the lines from the websites, and then try to compare them to chase steam in an automated way instead of sitting and watching sbrodds as the lines change. is this even practical? has anyone done this an have any advice?

i was thinking of having a table of each game for the day, and then separate tables for each books line history, but i think the db might get overloaded with that many tables? i was gonna use mysql

for example

table for 2/11/12

rotation	team
509	spurs
510	nets
...	...

then for each team, different tables for each books line changes:

2/11/12 pinnacle 509

time	team	spread	odds	total	odds
9:30am	spurs	-9.5	-109	194.5	-105
1:00pm	spurs	-10	-105	194.5	-105
...	...

2/11/12 5dimes 509

time	team	spread	odds	total	odds
9:30am	spurs	-9.5	-110	192	-110
3:00pm	spurs	-9.5	-107	194.5	-105
...	...

but lets say i'm scraping 5 books, thats 5 tables per team per day, lets say 100 teams, thats 500 new database tables per day..

anyone have opinions as to this structure?

don't do it like that. i don't know what the term "rotation" means, but essentially all you have are games you are wagering on, lines for those games at the different books at different times of day.

so you'd have a table called "games" and a table called "books," which should be self-explanatory. then a third table called "lines" which would have columns:

game_id (points to games table)
book_id (points to books table)
time
spread
odds
total
whatever else you need to described a specific wager at a specific time.

so 3 tables, one of which will hold most of your info and grow large. if space becomes an issue, you can write a script to archive or delete old rows, as it sounds like you only need them to find good bets? what does "chases steam" mean?

I don't think you fully understand normalisation, it's definitely a bad idea to have a new table for each team or game.

For example, start with a table of teams:

Teams:
ID
TeamName

Then a table for games:

Games
ID
TeamID
VersusTeamID
StartTime

Now in 2 tables you can store all the teams, and all the games. You now need to record the odds, simply do it this way imo:

Odds
GameID
DateAndTime
Spread
Odds

If you want to get odds from different bookies, add another table:

Bookies
ID
Name

Then extend your Odds table with BookieID

Odds
GameID
BookieID
DateAndTime
Spread
Odds

In MySQL this structure will be good for tens millions of records (space shouldn't be an issue imo), and for querying data in an easy manner. In regards to if this has been done before, yes it has! I met a professional football better in a gambling convention party thing and he has a team of about 10 people working for him fetching odds and making computer automated profitable bets, you would of course need to extend this database a lot to be able to make those calculations though. It's a very competitive and sophisticated market if you want to get into that.

thanks for your help so far. let me clarify some things

rotation number is basically the unique id of the team for that day. so there will be 501, 502, 503, 504, etc. usually team 501 will be vs 502. if you check out this site, you will see what i mean:
http://www.pinnaclesports.com/League...A/1/Lines.aspx

so in my example, 509 vs 510 is spurs vs nets. i didnt know if i should put in the full game as one row in the db or each individual team.




i'm not sure if this is the best way. "chase steam" means find websites (books) which have 'off' odds. so if the whole market has -9.5, and one book has -10, then thats an opportunity (albeit small)

but the whole market moves as a whole. so i was gonna input a new row into the db each time i scrape a new line. so in my example above, you can see that at pinnacle, the line moved at 1pm a half point, so i entered a new row. likewise at 3pm at 5dimes website, their line moved as well.

so if i put all this info in the "lines" (or gullianian's "Odds") table, then i dont know how i would be able to compare the lines at the books programmatically. so in my example, 5dimes was slow to move compared to pinnacle. so i could write a program to compare the last row of pinnacle (1pm, -10) and the last row at 5dimes (3pm, -9.5) and see the discrepancy, and send myself an alert somehow. if the "lines" table is just constantly adding rows to the bottom, i guess i could still compare the latest, id just have to learn the sql query to get the last pinnancle row and the last 5dimes row

actually if you check out this site:
http://www.sbrforum.com/betting-odds/

click on any of the odds and you will see the history, this is what i mean that the odds are constnatly changing throughout the day, so i want to be able to programmatically find the differences instead of sitting and staring at that site all day

when you have the bookieID column it's basically the same as having separate tables for the bookies as all you have to do is include a WHERE clause to limit results to your particular bookie. the benefit is that you can now select from the same table to get info about ALL the bookies simultaneously.

the sql query to get the most recent line from each book for each game is something like:

hrmm, great ballin4life, thanks that makes sense. my lack of knowlege of SQL means i had no idea that i could do MAX(date) to get the latest that way

Hey all,

I have put together my first iOS app and I am looking for some beta testers. Its a super simple weather app and I am looking for people who are able to test on the following devices:

iPod touch
iPhone 3Gs
iPhone 4s
iPhone 4

Testing probably wont take you more than about 5 minutes, I basically just want to make sure the app runs without crashing on all devices before releasing it (I only have an iPhone 4).

If you are interested, send me a PM and I will send you an itunes code. All I ask is that you open and close the app a couple of times, and email me just to confirm you didnt have any crashes (obv extra feedback would be much appreciated but is not necessary).

Thanks!

G

wooster,

emulators?

(i don't have an iDevice so i can't help you.)