Development ⇒ The future of TeXnicCenter

[planetmarshalluk]

I think that the first order of business should be to establish why development on TeXnicCenter is so slow. If it is the choice of toolset, then those issues can be addressed, but bear in mind that TC is an IDE for the Windows platform, and the toolset should be chosen accordingly.

Moving to a crossplatform GUI toolset is, IMHO, only worthwhile if the intention is to take TC cross platform. If it is not, then it makes little sense to choose WxWidgets or QT when the exisiting codebase is MS based and the tools for MS development are widely available.

Personally, my choice would be to keep TC Windows only, and gradually move to a .NET based architecture using .NET-MFC interoperation code a bit at a time. As a commercial .NET developer with Native C++ and Java experience I believe that this is the most productive avenue. Community developers can make use of the free IDEs to develop .NET code while those equipped with Pro versions of VS can provide interfacing code. This way TC can avail itself of the latest developments from MS such as WPF.

Regards,
Andrew.

[schroeder]

If you have a large base of C or C++ code, it makes sense to continue to use the same language instead of re-implementing everything in Java.

Moreover (and it is a more general advice), I warn you against coding in Java like you formerly coded in C++. It's the quickest way to make a slow and bloated application.

The majesty of theblues

Though...

The idea of changing to java was not because I think that java is the world's best language. The by far best thing I can see here is the availability of the Eclipse platform (to be more exact: the Rich Client Platform RCP). I do not know of any comparable framework for building integrated development environments (and TeXnicCenter is one) for any other programming language -- and that for free and as open source

...it seems to be a good idea, but only as far as it can guarantee that the new, java-eclipse-based TeXnicCenter2 is better than the current release, as for its usability, its ease of configuration, and the like.

Otherwise, the risk is it would simply turn out to be a "style exercise": wonderful as it may be, but not much more.

As a matter of fact, now - in my opinion - under Windows there is nothing more "functional" and useful than TeXnicCenter, if one is looking for a easily manageable Latex graphic front end, but - moreover - one which actually works (otherwise, the obvious alternative which comes to mind is to move to Linux).

So, should you finally decide to carry out the "java+eclipse migration", ensure that it really would be something more than a mere "style exercise". It should be a TeXnicCenter2 from the start, and should not need 10 or 20 intermediate releases for getting to the - let's call it this way - "breakeven point", or the "real-usefulness stage"...

Peace

[Sebi]

As a matter of fact, now - in my opinion - under Windows there is nothing more "functional" and useful than TeXnicCenter, if one is looking for a easily manageable Latex graphic front end, but - moreover - one which actually works (otherwise, the obvious alternative which comes to mind is to move to Linux).

Objection your honor!
WinEdt offers, due to its own macro language, by far more functions than Texniccenter (but, it is not free). Take, as an example, environment based formatting options, variable-width fonts, or "active strings".

Concerning linux, I have searched for quite a long time, but I cannot find any useful IDE that is comparable to TXC (Kile lacks online spellchecking, a very important feature).

[schroeder]

Objection your honor!

Overruled!

WinEdt offers, due to its own macro language, by far more functions than Texniccenter (but, it is not free)

...and, since for me free is definitely a plus... ^¿^

Anyway, I admit that, as I happen to do every once in a while, in my previous message I was generalizing a bit hastily. But I said clearly that it was only my opinion...

Take, as an example, environment based formatting options, variable-width fonts, or "active strings".

...hm... I suppose you mean variable width fonts in the editor.

Well... yes, I admit it could be useful, or at least aesthetically pleasant, to have an option for changing, even in the editor and every time you like, to variable-width fonts "on the fly" (and many other things which now I can hardly imagine). I have never felt the need of complaining about their lack, only because having monospace Courier (or any other fixed width font) in the editor doesn't hurt me that much...

But, as for giving WinEdt a chance, my problem, with this kind of try-before-you-buy tools, is that the time one usually has available before the program stops working is always (I repeat: for me) too little. So I usually tend to not even take them into consideration as viable alternatives to free tools. Once more, I admit this is a bias - mea culpa! - but, that's it...

Concerning linux, I have searched for quite a long time, but I cannot find any useful IDE that is comparable to TXC (Kile lacks online spellchecking, a very important feature).

Regarding Kile, I don't agree. Since I tend to use spellchecking very parsimoniously (i.e.: hardly ever), personally I don't consider the lack of online spellchecking in Kile to be a flaw.
But, as usual, nobody (and nothing) is perfect ;-D

As a final consideration, I'd like to rephrase what I was trying to say in my previous message: how much improvement is yet to / can be done on TeXnicCenter? Surely, a lot.
Actually I'm still waiting for a Latex graphic front end really capable, in case, of behaving like sort of an ultra-advanced word processor / desktop publishing engine, allowing the user to take advantage of the full power of Latex, with all its enormous typographical accuracy, without necessarily being compelled to learn the syntax of a "programming language". I think the time has come to begin to think that way, and to stop considering Latex as a "guru-only" matter: that is, just the same process Linux has gone through during the last years...

So, will (and can) these improvements come from joining TXC to the Eclipse platform, or from rewriting it in Java, or from whatever else? Well: Eclipse is well-known as a Java development environment, but apparently this is not its only power anymore. Now, for what I've learned in the last days, it works very well also as a C/C++ development environment (and, as far as I can see, it works really well). Which is what theblues was pointing out here...

So perhaps this strong integration "of the platform as a whole" might really allow to produce a much more powerful and much better TXC2 than the current one (which, as I repeat, in my opinion is already very good), without even the need of rewriting everything from scratch. Of course, I hope it will ;D

In any case, my best wishes to the author.... and to everybody else.

Peace

[elguero]

Hello everybody

I'm far from being a coder or having any significant knowledge about coding. But I know a little about the difficulties of platform independency. Recently I learned that Mono is a good platform-independent .NET extention. It's standard language is C#. Programs can be developed on NT and Unix aswell and work on both systems.

[polet82]

Why not to support the current TeXnicCenter, while finding the best ways to rewrite it?!
It is a common and well appreciated way of development.

There is a whole bunch of bugs floating around, waiting for a quick fix. Instead of proposing the solutions here in the forum (which is definitely appreciated), why not to make fixes?

I know that it is always fun to ruin everything and start from scratch, but that is not how the serious projects are developed. And considering a huge users base that TeXnicCenter already has, it is a serious project.

[Sven]

Why not to support the current TeXnicCenter, while finding the best ways to rewrite it?!
It is a common and well appreciated way of development.

[...]

Find people common with Visual C++ and MFC 6.0 knowledge (and licenses) that have the free time to do these fixes and implement further features and I fully agree.

In my daily business I'm the manager of a software development department with 15 people and believe me: I'm always against throwing away software and writing it new from the scratch -- something most (young) developers prefer when problems are coming up.

The problem with TeXnicCenter is, that it is open source. So everyone who will work on it will do it in his/her free time and without getting a cent for it. The problem with the current platform (Visual C++) is that the license costs some hundred dollars (please do not come up with the Express editions again -- they do not bring the MFC library TeXnicCenter is based on). Most people available for open source projects are working with current "hype" languages (mainly Java and C# for desktop applications). The people who have worked on TeXnicCenter in the past (mainly Tino and me) are no students any more since a few years and we do not have any time to do the work on the code and though we've started many tries to find new developers -- but we did not have success.

As mentioned in one of my first posts in this topic, the people are the only reason for thinking about another platform.

[polet82]

Yeah, MFC is no fun anymore, as it was 10 years ago.

This is quite strange that nobody wants to pick up such a successful project as this. There are a few business models for open source software development flying around. Ads/banners on the web site (I know it is evil, but still), donations may work, or even grants/sponsoring. Why don't you put an ad on the project face page, sourceforge and latex communities? The crowd is waiting for a new version.

Another question is how to recognize a good guy from a bad guy, who'd forget about the project in a week. You should have some experience with that, managing 15 goblins

[yostsa]

As an end user who just switched over to TeXnicCenter from LEd (due to unexplained crashes), I do hope that some continued development of TeXnicCenter will take place. You've got a great start on a very useful app, and I'd hate to see the progress end. Even if you do have to start from scratch to recruit developers, I am looking forward to whatever improvements you come up with.

[Sven]

There are a few business models for open source software development flying around. Ads/banners on the web site (I know it is evil, but still), donations may work, or even grants/sponsoring. Why don't you put an ad on the project face page, sourceforge and latex communities? The crowd is waiting for a new version.

Correct, the crowd is waiting ... but the crowd does not want to support actively. I've made TeXnicCenter open source some years ago to make it grow fast. But TeXnicCenter showed me, that the open source idea does not work for small software -- even not for a successfull software like TeXnicCenter. Besides Tino Weinkauf no one joined the project on a continuous time, though we have posted many requests on sourceforge and in the TeXnicCenter users list. Most people love open source -- but not because they can help to make the projects grow, but because they can be sure, that they wont to have to pay for it.

Regarding your "business models": I've tried exactly this some years ago, because I love TeXnicCenter and wanted to enhance it. I've offered professional services related to TeXnicCenter and LaTeX (including the creation of class files for company specific documents and things like this). And guess how many orders I have received? ... Right: 0!

Regarding the adds on the project site: They are already there and they hardly bring in enough money to finance the server the TeXnicCenter site and this community is hosted on (parts of this are still payed by me privately -- nice open source world [the ballance is good today, but in the first years of TeXnicCenter I have spend hundreds of dollar by my own to make it available to the peopel and I've never earned this money back]). Before I had a donation system for TeXnicCenter, but it was hard to distribute the donations among the developers, that were involved (how much money do you give someone that implemented only a small feature in contrast to someone that works on a continous base -- gues what: there was a lot of trouble about the donation issue and that's why I cancelled it) and the donated money was nice, but by far not enough to just even thinking about of doing this full time...

I think for some large business related open source software there may be a marked for professional services -- but not for small software like TeXnicCenter. And as long as there are no volunteers to work on this project, there will be no new version... Sorry for that