12 replies to this topic

[mattjmorrison]

Hello All, I just recently started a new open source project and I'd like to see if anyone would like to contribute. The project doesn't have specific purpose other than just accumulating a lot of common reusable code....any and all languages are welcome. If you would like to contribute, have some neat, fun or crazy code that you would like to share, or have questions or comments visit our Google group at: groups.google.com/group/software-developer-group. All are welcome! Hope to hear from some of you soon!

[WingedPanther]

What kind of license are you looking at (GPL 3.0 is very different from MIT)? Any particular languages?

[mattjmorrison]

Apache License 2.0. Any and all languages are welcome.

[MeTh0Dz]

Sounds like this project will fail as there is no real goal.

[mattjmorrison]

There is a goal...just not a specific purpose...or a specific use...Most open source projects have a specific purpose, some kind of framework, or utility that a lot of times buries potentially re-usable code...this project isn't for anything specifically...it can contain anything from very basic snippets to help beginners to advanced regular expressions that may solve a common problem that a lot of developers have. This is a project for polyglots. When developers need resources and/or code examples for multiple languages for a single project....they can come to our repository and it will all be there.

[MeTh0Dz]

Why on earth would they come to your repository when their are forums, databases, communities, etc. that are filled with all kinds of information by all kinds of "good" coders?

[mattjmorrison]

Because the repository will be a one stop shop, so to speak...why on earth would you use a web search engine to search through web pages of endless posts on forums, databases, communities etc when there is a repository full of source code. Why search web sites to look for code instead of searching a code repository to look for code....it seems to make sense.

[WingedPanther]

Unless there will be organization of the code by function/language in an easily searchable format, this isn't likely to be helpful. For example: statistical frameworks in C++, Delphi database connectivity tools, etc.

Also, from my experience the reusability of code is highly dependent on what you are doing. As an example, you may develop a framework for an application to be able to dynamically adjust SQL statements based on the target database. This is a really cool framework, but only useful in the context of the relevant databases that you deal with. To create such a framework that will effectively work with ALL possible databases will be a nightmare, however. OK, we can write it to work with the top dozen or so database platforms. That's helpful, as long as I'm not using one extra one. As a practical experiment, try to write a utility to extract a database's table structure and will return the results in a common format. Realize that Oracle uses different column types that MS SQL, for example.

[mattjmorrison]

But you're talking about a framework with a specific purpose...that's not what we're doing.

[MeTh0Dz]

Because what you are doing is a much worse idea.

You'd be better off doing like what WingedPanther said and writing a framework for something.

[WingedPanther]

That was my point: reusable code is created for a purpose. Truly reusable code ends up being something useable for everyone. That's how things like the Boost library got created: as a bunch of solutions to missing functionality.

Unless the code in the project is well organized, it will be challenging to promote it.

[mattjmorrison]

Here is the thing...frameworks aren't an end all solution...what frameworks do is a thing called speculative programming...where they write a bunch of code assuming that someone will need it....and you end up with something that works...however you've got ten times more code than you need. Also, since frameworks are a "black box" a lot of times you'll run into a situation where that framework supports most of what you need to do but not everything. You end up spending more time messing with the framework and trying to use it that you would have been able to develop your own solution from scratch in the same, or less, time. As far as the Boost library comment...that's fine if you're developing an application in C++ only...but what if you need Perl, Python, Java, SQL, HTML, CSS and anything else...for large projects developers need to be polyglots. That is where this project is focusing. We aren't going to focus on a specific language...different languages serve different purposes and most of the time using just one language is not good enough.

If you're not interested in contributing, that's fine, but its happening with or without you.

Thanks for the interest.