Lets design a program

[COUGAR]

I thought since so many people on this board talk about so many different programs and it seems nobody is exactly happy with what they have ( if it just had a tweek here a twench there) well you know what i mean. Personally i dont think it matters ( etc etc )numbers that is all you need. The main thing is how you push the bets through the windows...Discipline and Patience. There are 2 programs i use and they work for me, but ... IF... I could have the perfect program i would design it myself as i am sure anyone would, but time and know how( a big lacking gap in my brain, where is that on switch) well , i know there are a least 4 or more designers on this site and i thought it would be fun for us averege joes to create OUR PERFECT PROGRAM. May be we can get someone to design it for us and we can all pool the cost. Is that possible??....Hummmmmmm..

I will start and just be brief. I would just like the basic stuff on one screen 20 factors or so, pace, speed, class velocity, feet per sec( etc etc ) on one screen so i can see it well and print out on one sheet. The main thing i would like would be a SUPER DATABASE. The kind of data that woud find all the best combiations of winning factors. Lets say you had 30 factors to go through. The engine would seek out evey factor one by one and every combination up to the 30 combinations to find the best results for. ANY CLASSIFICTION you wont, not only the top horse but may be up to 5 horses deep in each factor. seek exactas, tri, and supers along with pic 3s seeking the best combinations of factors and the best ROI...

Anyone have any ideas????


COUGAR

[Speed Figure]

I'm working on having "my" own program made. I just need to find the right guy to do it.

[BillW]

Guys,

I hate to sound discouraging but, to contract out anything but the most trivial of jobs ( Excell/Access type stuff) would be quite expensive, unless you find a handicapper willing to do it gratis. Something as robust as the commercial programs you are familiar with have a few thousand man-hours put into them. Even at minimum wage, you are talking well into 5 figures. (BTW, Most engineers I know will not work for minimum wage :eek: )

You are probably better off buying a commercial package and contract the developer to do some desired mods for you.

Bill

[GameTheory]

I have often thought about an open-source handicapping program. Open source projects have thousands of man-hours in them, but no one seems to mind not getting paid. But that's probably because the effort is spread out over many people. However, I think the number of capable programmers who are also horseplayers is pretty small, and the number of those people who would be willing or able to contribute to such a project is infinitesimal. In other words, it would basically be a one-man project, and why would someone put hundreds of hours into something like that and then give it away?

[Amazin]

Cougar

BillW is right,but you have an interesting idea:A collectively built horseracing prediction program. As an alternative low budget program,I propose you start with a paper and pencil method. All members could contribute their input and then anyone could test it in the selections section. It could then get refined if it does not do well initially using member inputs for improvements.
You'd have to start like buiding a house,with foundation,framework and blueprint.

[BillW]

Originally posted by GameTheory
I have often thought about an open-source handicapping program. Open source projects have thousands of man-hours in them, but no one seems to mind not getting paid. But that's probably because the effort is spread out over many people. However, I think the number of capable programmers who are also horseplayers is pretty small, and the number of those people who would be willing or able to contribute to such a project is infinitesimal. In other words, it would basically be a one-man project, and why would someone put hundreds of hours into something like that and then give it away?

Most open source developers are refugees from the disfunctional corporate world where schedules are dictated by marketing depts. that have a simplistic view of technology and little recognition or concern for the complexity of a project of any size. In open source there is an opportunity for peer and user recognition and also the possibility for a bullet point or two on a resume.

In the handicapping world, these potential perks just don't exist. (no real world points for being a degenerate gambler :)) The user audience would require tons of support, without which the project would be doomed to failure. As you stated in another thread, without the software working to the users expectation, it doesn't matter how well it is done ... no style points either.

That being said, In a perfect world, I would love to be involved in a project like that also.:D

Bill

[COUGAR]

Call me crazy, but i just dont see it taking 1000+ hours for a project like this, but if so, one could use a skeliton shell of a program already in existanse to cut the man hours down. Then again, is it possible for say 20 or so people to agree on what they would wont in the program.? I am not sure, but i would hope to say yes, then again i am an extreme optimist. But lets say at most it took 400 hours at twenty five dollars an hour for one or multiple programs with interest as well and you had 20+ people in the creation.??

COUGAR

[cj]

Originally posted by GameTheory
... In other words, it would basically be a one-man project, and why would someone put hundreds of hours into something like that and then give it away?

I have, as a few on the board here can attest, but very limited. Its good to have objective eyes test your work. Also, it's human nature, we like to share and show our capabilities.

By the way, if you guys haven't checked out the power of GTs ChartGet, and especially the Parser, you don't know what you are missing!

[gonin02]

the parser has saved me so much time ! and you can customize the parsing to no end ! amazing tool...

Originally posted by cjmilkowski
I have, as a few on the board here can attest, but very limited. Its good to have objective eyes test your work. Also, it's human nature, we like to share and show our capabilities.

By the way, if you guys haven't checked out the power of GTs ChartGet, and especially the Parser, you don't know what you are missing!

[Brian Flewwelling]

Originally posted by GameTheory
I have often thought about an open-source handicapping program. Open source projects have thousands of man-hours in them, but no one seems to mind not getting paid. But that's probably because the effort is spread out over many people. However, I think the number of capable programmers who are also horseplayers is pretty small, and the number of those people who would be willing or able to contribute to such a project is infinitesimal. In other words, it would basically be a one-man project, and why would someone put hundreds of hours into something like that and then give it away?

GT, you are a programmer, right? Does this infinitesmal fraction include you?

It does include me. But two points

1) Don't let this proposal for a group project get out, or DickSchitz and DarrylFU might get all upset like they did last time i proposed the same.

2) the Value to me (as a programmer/handicapper) would be to see the "Factors" that several people could agree on.

I would suggest there are way more programmers who would be willing to work on this than there are Factors that we could get full agreement on :)

Fleww

[Brian Flewwelling]

Originally posted by COUGAR
Call me crazy, but i just dont see it taking 1000+ hours for a project like this, but if so, one could use a skeliton shell of a program already in existanse to cut the man hours down. Then again, is it possible for say 20 or so people to agree on what they would wont in the program.? I am not sure, but i would hope to say yes, then again i am an extreme optimist. But lets say at most it took 400 hours at twenty five dollars an hour for one or multiple programs with interest as well and you had 20+ people in the creation.??

COUGAR

You have a dream, and a good one i might add. But you don't have a lot of experience developing programs, or you are way faster than most. A 1000+ hours is not a BIG project!!


The skeleton shell is the easy part, the time is in the details. And that is where your suggestion has the most merit. Several databasers working on queries could speed the work along.

What you need is One, or Three, Managers to coordinate the project, and several worker bees to do the Code (including the queries) and a larger group of 'suggesters' who take the role of Primary Users. The 'suggesters' do just that, they make suggestions about User Interface and Questions they want answered.

Assuming this goes a little further: I suggest the group becomes mostly 'closed' once the project starts even tho the goal is to offer it Free. It would not be productive to have the steps exposed to the general population of (even) this board during the delicate early stages.

Maybe i am just hoping this goes a little further
:D :D :D

Further, if this is to work there is no room for Paying, it must be a group effort with all time being supplied gratis.

Fleww

[Larry Hamilton]

An interesting logic problem to determine who will participate:

There are two skill sets involved: handicapping (HC) and programming (Pgm)

HC has three levels of skill: "win", "lose, but" and "lose"

Pgm has three levels of skill: "can", "can, but" and "can't"


when you cross the matrices you get

can program and wins
can program but (fill in the blank) and wins
can't program and wins

can program and loses but (fill in the blank)
can program but (fill in the blank) and loses but (fill in the blank)
cant program and loses but (fill in the blank)

can program and loses
can program but (fill in the blank) and loses
cant program and loses

ok, let's see if you can determine who will stay and who will go.

All the "wins" have no reason to stay. What they are doing works and makes them money. (3 gone)

All the "Fill in the blank" are gone as this usually means they dont have the time to assist. (4 more gone)

What remains are two groups: 1) I can program but am a losing handicapper, and 2) I cannot program and am a losing handicapper. What one can conclude is that you have tools to digest the information but that all the information comes from unprofitable sources.

To succeed, you need a hook to attract some who win.

[Larry Hamilton]

Now it gets personal. Lets say that you have found a way to attract 2 winners to go with your 7 losers. The two winners have a clear proven way to profit. If you mix the two you MAY get more winners, but you may also turn them both into losers. On the other hand if you mix the ideas of either winner with 7 losers, guess what you get.....

Whats the solution? damnifiknow

[Larry Hamilton]

There's more. It's been bandied about that 5% win while 95% lose. On any given day, less than 100 individuals come thru here. As many lurk as post. That means that on any given day there are at most two winner in the daily group. How are you going to know who those two are and how to attract them?

[Larry Hamilton]

there is a solution, though it will be tedious. You can take the ideas of those who lose and cross that idea set with other's idea sets infinitely until you find pools of profit.

[Amazin]

Larry

This is uncharted territory.How many other horseracing programs do you know of were collectively produced on the internet. Your comments are premature.

[Larry Hamilton]

and yours are irrelavent as you have proven yourself illogical and unthinking

[Amazin]

You fly off the handle too fast. I stated a fact about a horseracing project and you decide to get personal. People like that usually need to seek a therapist and their statements are suspect due to their mental condition.

[ceejay]

Just a view from an observer who is not likely a partisipant in the consortium.

I don't think that open-source and free is the appropriate model, at least at the beginning. Those who invest their ideas, time, and/or money should be rewarded in a success case. Perhaps, making source code (with distribution restrictions) available to the group is a better plan so the consortium members can personallize the applications.

I have been involved in several consortiums (not in the software field) in the past where after a propritary period of time the work becomes public-domain.

[Dave Schwartz]

Please permit me to jump in here and add my two cents worth.

First, I am certainly not a "contender" for this project. I simply do not have the time. But I have been at this programming thing for quite awhile and I think I have some good stuff to add.

If I WERE involved in this project, I would suggest that a "board of direction" (BOD) needs to be appointed. This would be a set of core users that control the project's development.

The BOD should be comprised of the people most committed to the project in terms of time, money and expertise. And make no mistake... there will be money needs. Perhaps the programmer needs a more powerful system in month 5 or the group decides to start with a RAID database system. Or perhaps some software needs to be purchased to fill a need rather than developing it from scratch. (You should spend some money and outsource whenever possible IMHO.)

My point is that it all starts with a central core of guys that are willing to commit... something significant.

At HSH we have such a group, although many people aren't even aware of it. They think it is all me deciding which way to go but it isn't. You should try to get the BOD together at least once per year in person (as we do) but with communication as it is today you could do it all by email. (And, BTW, we invite the users to MOST of the public meetings.)

Now, understand that the purpose is not to keep everyone else in the dark. Your BOD will be constantly bombarded with ideas for direction, especially in the beginning as you decide what kind of program you want to write. Some of the ideas will be downright brilliant while others will border on stupidity. And sometimes you can't tell the two apart.

There must be some kind of "triage" procedure that builds a development agenda. For us, the agenda position of a new idea is determined by how important the feature is to the long-term direction of our software, how much MOST users (not just the BOD) would benefit by it, and how difficult it is to implement.

One idea that you might consider to start is to get PA to set up a private forum that only people involved in this project have access to. Of course, first you must HAVE a project leadership team in place to decide who is who. Another alternative is a private email list.

I hope this helps some. Since a commercial developer such as myself has no place in this project, I will not post again unless invited to. I have no desire to horn in on your project nor to profit from it. I wish you all well in this endeavor.



Regards,
Dave Schwartz

PS: And one more thing... Don't forget that PA is first and foremost, a programmer! He understands this stuff very well.

[COUGAR]

Boy oh boy.... What a mess. Its starting to sound like my old days of playing BJ. You have the mathamaticions ( excuse my spelling, WHERES THAT DAMN SPELL CHECKER) on one side who were great at theorizing and running simms but most couldent really play in the real world ,and then you had the players that could actually make money in the field. I am not a math guy never have been never will be. My tactics went againts what most math purists would agree on because i did not like the WILD RIDE SWINGS..... Anyway long story and you know what i mean.

It souunds like you guys are trying to create the BLACK BOX, theres no such animal, never will be. The funny thing is.... IF there was a BLACK BOX the average joe , that 95% would F___K it up anyway and still lose because of no money management skills, patience and discipline day in and day out.. Actually a person that trully posesses thoes skill ( and most will disagree) can actually overcome a small disadvantage in a negitive expectancy game.. Somthing like craps or baccarat were only the best 1% diadvantaged bets are being made..

My main goal would be to develop a super powerful database where one could search past historal racing data in truly unique way, like has never been done ... At least that i know of.....

Sorry again for some spelling errors, that was never my strong suit...

COUGAR

[GameTheory]

I don't handicapping skill is very important, actually. Handicapping ideas are, however. You don't have to have the skill to use pace figures to make money to know how to compute them. I doubt most authors of handicapping programs can make a profit at the track. I think it is almost irrelevant.

Anyway, my suggestion would be to:

-- first, design a basic import engine (choose a default datafile, like $.50 TSN files) and database structure

-- design a basic display grid/export system

-- Use an existing free database library/system for database management. Berkeley DB or Tsunami Record Manager would be a good choice. Note that use of one of these might REQUIRE the project to be open-source (at the least the "core") if you want to use them for free.

-- That's the basic core -- no handicapping stuff even in it. Just a way to import/manage/display/export data.

-- One more important part: design it in such a way so that all other future functions, handicapping, support for other data files, or whatever can be created as "plug-in" modules. These plug-ins could be written by anyone, and could be sold or could be free. The main engine as described above, however, would be free, and would have an API/protocol that the plug-ins would use to access the data, manipulate it however it liked, and then return the processed data back to the main engine for display or export to the user.


Brian -- yeah I'm a programmer. In case you didn't realize it, I'm the guy from the racemath list that helped you with the archive stuff a while back. But, like Dave, I doubt I would have much time to devote to such a project, and certainly couldn't be the main guy. I probably couldn't do much more than offer suggestions and maybe test stuff out. I think chances of getting such a project off the ground are slim, and will take someone to step up and put a lot of work in without getting much in the way of reward.

[Larry Hamilton]

I didnt mean to cause trouble, Ill butt out

[Speed Figure]

Cougar,

You should just do "your" own program. Forget about everyone else.

[COUGAR]

Speed Figure,
I realize that...Always have, but i thought it would be fun. It should be, life is tooo short, and i would never expect anything for gratis...

Cougar

[DJofSD]

I find this to be a very interesting idea/

Lots and lots of questions though.

And as I often wont to do, let me ask a rather point couple: what programming language would be used and what would be used as the underlaying data base management software?

The devil is in the details, guys.

DJofSD

[Brian Flewwelling]

Originally posted by DJofSD
I find this to be a very interesting idea/

Lots and lots of questions though.

And as I often wont to do, let me ask a rather point couple: what programming language would be used and what would be used as the underlaying data base management software?

The devil is in the details, guys.

DJofSD

This sounds like a tentative probe by a potential coder :)

If i were to be involved, the language would need to be VisBasic, and it will work with Access Data files quite seemlessly. The user would not need to buy Access to use the program, but if he had it could get at the data independently of the program.

Fleww

[DJofSD]

I've done some programing in Visual Basic - no thanks. Access, again, no thanks.

Most DBMS software including the Jet engine allows for the free distribution of the core modules to allow the applications using the engine to work. You just can't ship/install the whole DBMS itself.

DJofSD

[Brian Flewwelling]

Originally posted by DJofSD
I've done some programing in Visual Basic - no thanks. Access, again, no thanks.

Most DBMS software including the Jet engine allows for the free distribution of the core modules to allow the applications using the engine to work. You just can't ship/install the whole DBMS itself.

DJofSD

Your message seems incomplete. It doesn't mention the language or the DBMS you would prefer to use for this project.

Fleww

[DJofSD]

What I would use for myself is very likely not what most folks would use.

I personally code PC apps using Delphi. The built in DBMS that's basically the old Paradox engine is good enough for me but there are other replacement products like Apollo that could be substituted.

DJofSD

[pmd62ndst]

I was going to stay out of this thread until I read the DJofSD uses Delphi. What about using Interbase for a database?

I'm a former Borland employee who codes in Delphi and uses Interbase 6.0 Open Source version (free) for my homegrown handicapping app.

Borland doesn't offer the installs for that version of Interbase anymore but I've got it stashed away and I'll be willing to donate it along with any coding help I could contribute.

PMD

[Amazin]

Well I can see this project is moving right along. I previously suggested that instead of a colllosal computer program, we start with a simple paper and pencil method. Any suggestions? Don't all rush at once.I know you're all so enthusiastic here. Maybe it could get reviewed by Phillips Racing Newsletter,or get sold by RPM and we'll be rich I tell ya, rich! (That's supposed to be a joke) .