Looking for someone to create/modify software for this forum [5500+ BTC]

[M4v3R]

Starting in March would be fine, but I want to make a decision about who the bidder will be by the end of January. (If there are no great bidders, I will probably abandon this bidding process, start an open source software project, write the forum core myself, and place bounties on all other required software aspects.)

By the way - this seems VERY reasonable for me, and it probably would end up better than placing it all in one hands.

[theymos]

I'm still looking through the submitted bids (a lot of material was sent to me privately, and I've been busy). I'll continue to accept bids until I've put a significant amount of time into writing the forum core in anticipation of that alternative development method I mentioned. To be safe, anyone interested in doing the whole job should send me at least a basic bid ASAP.

[CIYAM]

I know that this is getting off-topic, but this really bugs me. Software engineering is more than just writing and debugging code, you need to be able to design systems and document them too. Programming is the easy bit, design and documentation are IMO much harder problems, and if you aren't documenting then you're missing out on valuable experience in your journey as a developer.

I do understand the value of system design and documentation - in fact when I release my software platform you might be rather surprised to see that in fact that what you look at and edit with is basically a design document (source code for apps is 100% generated from a sophisticated model that includes specifications that describe in an "intentional" manner all objects and functions of the software).

You might want to read up about what Charles Simonyi has been working on for at least the last 10 years to understand the kind of paradigm shift in software engineering I am talking about.


Cheers,

Ian.

[LehmanSister]

- An API or a NNTP interface.

I think that as many, many, many forum admins have done time and time again... You've just reinvented the functionality request of the uucp era internet, except with JavaScript instead of curses... Besides, everyone's good loot is in their browser, not their shell anymore. NNTP is great.

The daemons have handled a lot of abuse, and continue to. I believe someone created an "alt.bitcoin.test" already. The protocol is purely peer to peer, prima facie.

Hashcash to pull subjects and get content, coin or barter to post and be heard beyond your direct peers. Minimal cost for recycling of large chunks of previously delivered recent text along the path (quotes). The idea of focusing on HTTP is kinda silly, otherwise we might as well call it Layer 8. The should be the focus of a good protocol, not the display (imho).

[Xenland]

Maybe we(as in community) can use this forum I created http://pipboy3000.com/social and I'll release the forums source-code on Github and we can all contribute and then the funds will be distributed based on the most code contributed?

Its code is very very early but its already well commented and isn't complex it should be easy to contribute upon with little effort.

I think we can all come together and agree on how payments will work as far as contributions go, I was thinking for code distribution to be completely fair I would get paid at a fixed rate of 10% of the total payout that would also mean no matter how much code I contributed I would get paid no more or no less. The incentive for me to contribute greatly to the project will be of course the 10% as well as the success of the forums software which would also increase security, functionality and useably for my projects that will incorporate this forum software and also I would like to see a secure forum and done correctly for once.

Everything is negotiable, its just a thought. pm me for more details to reduce spam on this thread thanks!

[daybyter]

You want PHP? So I would try to generate as much code as possible and use a RAD tool like CakePHP.

Determine what the user should/could do here in this forum and create a nice use case diagram.

Do the next step from there and create a data model. I would recommend ArgoUML with the data modeling profile, since it freely availabe and everyone could download and contribute.

Generate sql from the model (I've to admit here, that the available db modules are a bit dated and might have to be modified, but that shouldn't pose a big issue, since everything is Opensource and pretty straightforward).

Once the sql is there, create a database and feed the mysql into it to get a running database.

Install Cake and point it to the existing database and let in create model, controller and views for each table.

[Skybuck]

Have you considered PHPBB ?

[neo_rage]

Really, SMF 2.0 + some mods will solve your problems.

[Red Emerald]

Really, SMF 2.0 + some mods will solve your problems.

Then write it.

I don't think people appreciate how complex forum software is.  I've been working on it for a few weeks (still have to actually put a bid in), and it is not at all easy or simple.

The forum has categories which have boards which can have any number of child boards. All the boards have topics which can also be polls and all of those have posts.  The SQL for loading all of that is difficult enough and that's without any sort of permissions system.  Then theres also all of the stats collecting and personal messages to code, too.

[daybyter]

That's why I proposed a RAD tool. Don't write all the SQL queries. Let the tool handle it. Ok, the recursive forums have to be taken care of. Just put a parent_id in any forum, so it could become a subforum.

I have similar models here, if you want any help...

[Omni]

Have you considered MyBB? Imo it is the best forum software.

[CIYAM]

Seems a lot of people on the forum know very little about *modern* C++.

I have not had a buffer overflow (or even a memory leak) in my code for years.

[theymos]

Status:

Of the bids that have been formally submitted, I'm only considering Request's seriously at this point. If you think I didn't give your bid a fair chance (especially if I don't mention you below), PM me with more details.

- nhodges: I sent you a PM requesting more details but didn't get a response as far as I know.
- Red Emerald's work sounds promising, but he hasn't given me enough detail yet.
- Matthew N. Wright hasn't sent me a bid yet.
- BinoX hasn't sent me a bid yet
- danieldaniel isn't qualified.
- I'm uncomfortable with using CIYAM Pty. Ltd.'s experimental (in my mind) code-generating software for something where security and performance are important.
- windowsdefender probably isn't qualified.
- Maybe Xenland or FlipPro could do it, but I don't feel confident enough to give either of them the job.

There's now 1700+ BTC in the forum fund, and I'm still accepting bids.

I've been too busy (and unmotivated) to do any coding, though I've been thinking about the design a lot. I think it would work well to put all of the business logic into the database. PostgreSQL allows you to create new database functions using any of several languages (including C++, PHP, Perl, and Python). This design would be very modular, easy to extend, and easy to break into parts for many people to work on. Making many different front-ends would also be easy, and the front-ends could be "stupid", working only with views covered by many triggers/rules. I'm not a very experienced programmer, though -- what do guys you think of this design?

These two requirements can't both be achieved - unfortunately these are the two most 'accident prone' languages in wide use... making security nearly impossible.

Bitcoin uses C++. Hopefully Bitcoin is secure...

I don't find security to be very difficult on either. You do need to know what you're doing, but this is a prerequisite to winning this bid anyway.

[CIYAM]

Just to let you guys know that although not concerned about getting the contract (and I understand the reservations about my system) I have made some progress creating a "forum package" for my software system (which of course has *zero* manual code).

Although far from complete it handles ordering latest post topics first (after sticky topics which are also in latest post ordering) and allows for topics to be locked and moved. Child boards work correctly with stats totals updating to parents.

http://www.ciyam.com/forum7.png


Cheers,

Ian.

[perlboy]

Heya,

At the risk of looking like newbie that should just mind your own business I'd happily assist extending a vBulletin install to meet your additional requirements free of charge and for the good of this community. vB itself is ~$800 and is scalable and secure. I personally don't think a custom forum software is really worthwhile nor more secure.

Just my 2c,

Stu

[bitplane]

Seems a lot of people on the forum know very little about *modern* C++.

I have not had a buffer overflow (or even a memory leak) in my code for years.

Right now there's a critical vulnerability in the Bitcoin client, which just goes to show how easy it is to make an exploitable mistake in C++.

Also from a philosophical stance, if you knew your code had a buffer overflow then you would have fixed it. Malicious developers aside, security vulnerabilities are always unknowns. All code has bugs and some bugs are critical. You can only reduce your areas of risk, and in a language like C++ every pointer is a potential risk.

[CIYAM]

Of course you can write some terrible stuff in C++ but typically the only problems I tend to find these days are dealing with C style interfaces which are the ones most likely to need to deal with pointers to buffers (and even worse void pointers).

I am not sure exactly what the bitcoin-qt problem is *but* there is no such problem in the older client (so why not blame Qt rather than C++ itself?).

To offer a parallel criticism I remember years ago when Adaptive Server Anywhere first decided to rewrite their C++ SQL query tool in Java (the reason why I could never actually work out as the C++ worked extremely well). The resulting program was such an embarrassment (queries that used to take seconds taking minutes, exceptions being thrown all the time and memory usage ridiculously high) that they had to include the old C++ program anyway (and I don't think anyone would have seriously tried using the Java app more than once or twice).

So can we agree that "crap can be written in any language" and stop with the language bashing?

I won't dispute that C++ has a large (maybe the largest) learning curve but if you learn it properly then I don't see it being more likely to result in poorer quality code.

In modern C++ one rarely uses plain pointers excepting the case of using them for the purpose of adding extra "optional" function arguments. If you don't know about them please read up on "smart pointers".

[CIYAM]

Potentially critical vulnerability in versions 0.5 to 0.6 on Windows only.

you did notice this didn't you?

[bitplane]

So can we agree that "crap can be written in any language" and stop with the language bashing?

Basic probability says that the more lines of code there are, the greater the chance of it having a bug. As you add lines the chance of software having a bug approaches 100% and this applies to all code. Not badly written code, not code without unit tests, not code written by beginners. All code. For shipped code without unit tests and a dedicated QA team it's about 20 lines, for the best team it's about 100, for the best practices in the industry it's about 300. Does your codebase contain more than 300 lines? I don't care how good you are, even if you're Donald Knuth your code contains bugs.

Some languages are more powerful than others and the bugs are more dangerous. C++ is a powerful language where bugs have a high chance of being loose cannons.

In modern C++ one rarely uses plain pointers excepting the case of using them for the purpose of adding extra "optional" function arguments. If you don't know about them please read up on "smart pointers".

No need to patronise me. Smart pointers prevent dangling pointers, which is only one class of security problem with direct memory access. If you aren't using C++ in a way that is potentially dangerous then C++ isn't the right tool for the job.

[Red Emerald]

If you aren't using C++ in a way that is potentially dangerous then C++ isn't the right tool for the job.

Haha. I like this.