what are you trying to do?

I am just trying to write a very small script for calculating (re)shoving ranges.

I'd like to be able to input my own hand and a range for my opponent (e.g. top 15% of hands) and then inputting these in Pokerstove to get my equity and use this in my script.

So, specifically, given my hand and my opponents range, I'd like Matlab to call Pokerstove (or PPT or whatever) and get my equity.

If this is more of a programming exercise for you then the following won't help if you just want to get (complicated) Omaha EV-calculations done you can sign up at whats-my-line.com it's in beta but should be able to handle this. If it's of interest just send me a pm after signup and I'll unlock your account.

I'm fully frustrated with the work I'm doing right now. I think reading too much here and other odd stuff has made me excessively paranoid about finding someone to do the company website. Of course, I'm somehow in charge of this garbage. I spent a few hours today collecting my thoughts and writing it all down. If I'm on the wrong track, let me know. Writing it out helps me to focus my thoughts and see the issue. Writing this also revealed to me a few ideas that I didn't notice.

Happy reading. If it's TL;DR, then don't read it. There's no summary. I do want to point out that web dev companies really make this whole process far harder than it has to be. Maybe if you are looking to build / expand / start your own, this will help you out.

I'm going to have a long week ahead of me.

Have you considered finding company websites similar to what you are looking for and contacting them to ask who developed their website and if they would recommend them? Recommendations are really the only way to find a decent web design company.

Any of you guys on Geeklist (http://geekli.st)? I actually quite like it. Interesting to see what projects people have worked on.

This would be pretty uncomfortable. I wouldn't have any choice but to call our direct competitors and ask them.

There is one other thing though. I did whois search on certain companies and found that a few of them share the same IP address as some of our customers. This either indicates that our customers are owned by our competitor or it means that our customers have the same developer and are all on the same shared hosting plan. The second option is far more likely especially since most of these companies are all located within a few miles of each other.

Not on this, but it looks like it would be a helpful resource.

I'm going to totally whore out my company. We're not in LA, but we do work globally. I'm not a sales guy, but I can give you a contact number. Our website blows, but we do good work.

PM me if you're interested (PS, please don't ban me)

I dont know what business you are in, but I dont really see a big issue with contacting competitors to ask them who built their site. They dont need to know who is asking.

Also, unless you are looking for something very specific, it doesnt need to be direct competitors, just websites that have good design.

Eh...

There is *one thing* that we need implemented that other companies have that is only partly implemented and I thought a lot about why this is the case. Imagine a function that should be available for all products, but in order to implement the function, each product must be manually loaded and the end-user has to do be able to do some manipulation with it.

One company recently came out that created the above service and sells it to everyone in the industry, and from what I have seen, the UI of each site that has it is drastically different and in many cases, impossible to use.

So, there's the following issues:

1- The company that created this program made it so that anyone can load it on the site, but it's not abstracted enough to be generalized, thus the companies that use this service have some products with the feature, but most of the products aren't available, thus the companies that have it don't have the time or knowledge to fully implement the program. Imagine a site has 1000 products in 7 colors each and when they first got this program, they had it available for 300 products in one color each. Something makes it too hard or too time-consuming to expand this into the full product line. After using the front-end myself, I believe I see where the problem lies.

2- There is what I think is a simple open-source solution to this issue. The open source program looks more like industry-standard UI than the paid-for service. The open-source solution also allows end-users to save as PDF, PNG and other formats, which may or may not be available on the paid service (I guess add $$$). The open-source solution also uses SVG, which is a huge plus.

3- The difficulty is determining exactly how to implement this program, and the paid-for program doesn't do a good job of creating a basic standard that everyone can model on. Solutions include:

* Drag-and-drop which I don't like.
* Simple link that talks to the back-end and loads the required photos from some photo file / database.

I believe all the above issues can be easily solved if the program is properly generalized and abstracted. Yes, the website maintainer will probably have to follow some instructions and open something scary, but I don't think the instructions will be too difficult. I think that the paid-for service was attempting to make it so that it was intuitive and easy for anyone to use it, but at a huge expense of fast implementation thus this is why this amazing and needed idea has yet to storm the industry. I believe that this is a huge first-move advantage for anyone who fully implements this service, and I believe it is worth the sacrifice in back-end GUI to ensure we are ahead of the curve on this, especially thinking that the current industry program doesn't seem to be working out for anyone anyways. Demand for this particular feature is very strong and it shows how bad the current solution on the market is if the only available option is struggling to get companies to use it.

Seems like a killer feature with first mover advantage should be developed in house. I take it this project is something you really do not want to do?

No, not something I want to do at all. If push came to shove, I would, but I would hope a semi-competent developer can do it. I also don't have the time.

And if I did it, the back-end would be for speed and efficiency, not usability for the non-coder, which is pretty much how everything I currently create functions. I put some thought into how something like this would be created. I'm thinking of creating it at home for giggles. The program might come in handy one day.

This just scares that crap out of me

I'm finally hacking together my first rails app. Say I have a restaurant model and I want to give each restaurant a cuisine. What are the best practices here? I know I could just give Restaurant a string category attribute (column? field?), but that feels like poor design. Should I be creating a Cuisine model and then storing an id or string value for a cuisine object (are there built in functions to handle this?). What If I want Restaurant objects that can have to multiple cuisines. What's the best way to do that?

MrWooster,

Site looks good (though I'm not sure I'd ever use it as I would be uncomfortable posting my session histories on the internet), but I do have a couple of suggestions:

- I think you can remove the header for each hand - there's no need to have "Poker Stars $3.32+$0.18 No Limit Hold'em Tournament - t10/t20 Blinds - 2 players" repeated every hand (not to mention 3 lines of vertical space). I think you could put the blind sizes beside the stack sizes.

- For heads up, I don't think you need to include the (2 players) for postflop.

- Personally I find your current cards hard to read and would prefer something bigger that resembles actual cards.

- I would reorder the flops and hands so that the the cards are descending in value.

- I think you need less horizontal space between the comments and the actual hand (or more vertical space between each hand).

- I think you can add hand counts to the display filters.

- On a less nit picky note: I think you could add a list of hands along the side that just display a bit of information about them (pot size, who won, etc), which scroll you to the hand when clicked on.

+1 for 99% of cases that is the exact wrong philosophy

Design is the most important thing. Building front end web-apps for speed and efficiency is a lot easier to justify imo

Thanks for the feedback.

This is a fair enough point. One of the features I am working on is the ability to have 'private' sessions. These will only be accessible by designated users. The idea being that you post a session and only give access to your friends/coach etc.

This is a product of the hand converter. My intention is to eventually implement a custom hand converter which would give me more control over how the hands are displayed but at the moment I use the deuces cracked hand converter. A lot of post processing goes on to determine extra information about the hands, but its not the prettiest code and I would rather wait until I have my own hand converter before making these changes.

Incidentally, if anyone knows of an open source hand converter please ping me

Personally, I have got used to the 'icon' style card which is very similar to the ones used on 2p2. Would not be too hard to add an option to change the cards though.
Interesting. I have not seen any converters that do this, but I can see why it would make the flop easier to read.

In theory, the columns should resize to fit the screen size, but unfortunately css is not my forte. Will have a look at improving this.

Could you explain what you mean by this?
Would really like to do this, but again, will have to wait for a decent hand converter.


Thanks again, really appreciate you taking the time to give some feedback

@daveT from what you are saying, if you can get the feature implemented you will be way ahead. If you have the time, then you could always develop it in house and then sell it to other companies later on down the line.

agree. at the very least you want whatever Ruby folks use for an enum.

cuisine should be a model -- what else would it be? how else would you edit cuisines, create new cuisines, etc.

i'm confident rails has a way to connect models together. something with attributes? i distinctly remember being able to set up "has_one" and "has_many" relationships. Restaurants obviously Have Many Cuisines.

Thanks, not sure why I was having trouble last night, but your post inspired me to use some new terms for a google search and came up with: http://guides.rubyonrails.org/association_basics.html which appears to contain everything I need.

there is this.. http://****************/apps/mediawik...itle=Main_Page


In your display filters show how many hands end in preflop, how many on the flop, etc just so the user can get a better grasp of the effect of selecting a filter.

Good idea regarding the filters Will take a look at the FPDB converter. I remember checking it out some time ago but havent had a look at it recently.

We use fpdb as the backend for Pokeit. While it's good for parsing hands and putting them into a database, it doesn't give you a way to display them on a webpage out of the box. If you're looking to develop a custom look and don't mind writing some of it from scratch, fpdb might be the right choice. fpdb is always looking for new devs to contribute so if you do decide to check it out, just let me know if you have any questions. Also, you can always hop on the #fpdb IRC chat or shoot an email to the fpdb mailing list fpdb-main@lists.****************.

Hey guys, anyone know if people who get your app through ad-hoc distribution can review it in the app store? I know in most cases you have to purchase it before you can review it.