... and certainly there's no one yet that can look at the data of a Tensor and say: Hey , I know what this ANN does.

Auto completion can't do it all. Well , of course not , but it's a great tool when you are learning a new library. Even after you have learned it's a big help. In a compiled language, the compiler might catch an error - but not a bug. In a compiled language the compiler will catch all syntax errors. And it will spot some bugs ( dead code, unused variables , uninitialized variables). If either of these routes is unsuitable, I would suggest finding an alternate language for your problems. Of course they don't suite my needs, I'll be using typescript whenever I can unluckily for some cases there is no other choice than using javascript.

I've done some typescript and I absolutely love what Anders Hedjelsberg did. The library I will be using still needs a typescript facade though , so I have to stick with javascript. I can't help but wonder what were the twisted minds that created this devil-spawn language thinking when they were going through the design process.

I still can believe the web runs on it! If there is any proof the devil exists javascript must be it. It is so easy to make a mistake and not catch it until you are debugging.

I'm not quite sure this is the appropiate place to start a discussion on this topic, since most posts seem to be related to programming. I personally find the metro interface to be quite ugly even for a touch based interface. If I compare it with my android's phone UI I find it lacks some rather basic features : 1. The ability to sort icons alphabetically , by frequency of use or according to an order I designate. 2. A quick way to switch / launch the apps I recently launched since I turned on my phone. 3. A Status bar 4. The ubiquitous navigation buttons at the bottom of the phone. Do you find the metro UI appealing?

Mycroft Holmes wrote:

I predict that almost no one will actually pay money for it if/when you do publish (remember we all built our own).

Hmmm... I was thinking in selling it at a VERY affordable price ( e.g USD $20 - 25 , trainning videos included ). My bet is that it will draw the attention of novice programmers, but mostly it will be done just for the experience of getting a comercial product on-line. Once again , thanks for the feedback.

Thanks for the feedback. One of my thoughts was to polish it enough so that even a non-programmer could use it ( but that is an altogether different creature which would probably take months to complete ). As-is it would probably be usefull only for novice programmers ( < 1 year experience ).Anyhow , I started this project because the company I'm in makes everything by hand and I couldn't stand the tedium of writting n-tier apps by hand.

Easiest way : response.redirect("MyOtherPage.aspx"); if you need to pass data you can use the session object ( e.g Session["firstPage"] = myObject) and retrieve it from the second page ( many programmers see the session as the root-of-all-evils, but I think its ok to use it to pass small( <10k ) object).

I am working in a c# code generator project. This first version was mostly hacked out in a couple of weeks. ;P , Now I am having qualms , regarding if I should publish it or simply keep it for myself ( I mean , not out of selfishness , but due to the fact that a comercial product needs a lot more work than an only-for-your-eyes tool, plus I get the impression that code generators are not very popular these days ). Currently the tool generates the following : Stored procedures for the tables (SQL Server only) Business objects Data Access Layer A basic navigation menu A list view for the objects. Any feedback will be greatly appreciated.