Overview
Examples
Screenshots
Comparisons
Applications
Download
Documentation
Tutorials
Bazaar
Status & Roadmap
FAQ
Authors & License
Forums
Funding Ultimate++
Search on this site
Search in forums












SourceForge.net Logo
Home » Community » Coffee corner » What next?
What next? [message #45622] Thu, 17 December 2015 15:54 Go to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
OK, so U++ is now better than ever, but still largely ignored.

Which is sort of understandable, because nobody knows and most of those that know do not understand (rest are happy U++ users Smile ).

As I do not see the point in developing complex framework/ide for tens of users, I would like to concentrate a bit more on PR in the next period.

Things are not that simple, number of public arenas open to "subvertising" of open-source software is now limited. E.g. annoucement post on google groups is now read by about 10 people total. Readmeat does not exist anymore.

So I am trying to figure out what to do next.

- I believe I have to write a document about U++ Core and its performance. I guess this one of early detractors from U++ and it should be properly explained why we need (and everybody should) avoid C++ standard library.

- The only suitable medium for PR I have found is CodeProject, where you can still submit articles about more or less anything. It is not so easy, because articles can be rejected. Anyway, I would start with version of reference/GuiWebCrawler.

- Maybe release some parts of U++ as separate libraries. E.g. memory allocator is a good candidate IMO.

- Maybe create core conventional derivative code editor from theide

Any other ideas or comments?

[Updated on: Thu, 17 December 2015 16:14]

Report message to a moderator

Re: What next? [message #45626 is a reply to message #45622] Thu, 17 December 2015 17:25 Go to previous messageGo to next message
Tom1
Messages: 1212
Registered: March 2007
Senior Contributor
Hi,

I wish I was better on marketing -- for a number of reasons. Maybe a short survey of what made us (tens of happy U++ users) make the switch to U++ could highlight the key selling points of U++, and then emphasize those points in marketing efforts.

Here's my story: I had worked years with Borland, then Watcom and finally Microsoft C++ tools. I was continuously looking (some heavy googling) for a solution to get rid of my dependence on Microsoft and their policies, (which by the way have not improved over the years, IMO). I wanted to find a suitably licensed (BSD) cross platform development solution that covered Windows, Linux and maybe *BSD. Finally 2006 I found U++ and slowly started to learn and rewrite my first application on U++. Step by step I became convinced of the superiority of U++ and started to realize how much more it was than what I had even dreamed of.

So, the initial key selling points for me were the BSD license and the cross platform development capability.

Another thing that comes to mind are the programming courses in universities, which could be a great advertising platform. A small project on U++ might give the spark. This would obviously require support from the professors.

Best regards,

Tom
Re: What next? [message #45628 is a reply to message #45622] Thu, 17 December 2015 17:56 Go to previous messageGo to next message
mr_ped is currently offline  mr_ped
Messages: 825
Registered: November 2005
Location: Czech Republic - Praha
Experienced Contributor
Maybe... (just thinking loud) one possibility may be to pick up some popular unmaintained open source project, and write alternative replacement in U++ from scratch. But I have no idea about what project that could be, maybe some front end cross platform client for some open messaging API, like jabber/gtalk, including video calls, etc? That may have impact over larger user base, but looks like lot of work just to get it going, and without heavy maintenance it would probably die soon after.

And with most ideas it would need serious involvement of some developers, the most probable outcome is that there will not enough people to develop and maintain it.

But in case somebody has some great idea about some SW, which can have serious impact and maybe is eventually willing to do development, write it here.

I personally hardly manage my own projects, lately I don't any programming outside of work, so this is just an idea from me, nothing more. :/
Re: What next? [message #45630 is a reply to message #45628] Thu, 17 December 2015 18:16 Go to previous messageGo to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
mr_ped wrote on Thu, 17 December 2015 17:56
Maybe... (just thinking loud) one possibility may be to pick up some popular unmaintained open source project, and write alternative replacement in U++ from scratch. But I have no idea about what project that could be, maybe some front end cross platform client for some open messaging API, like jabber/gtalk, including video calls, etc? That may have impact over larger user base, but looks like lot of work just to get it going, and without heavy maintenance it would probably die soon after.


Well, I would say that would end exactly the same - good but ignored. With PR, it could end better, but if we knew how to do PR then, we can do the same right now, right?
Re: What next? [message #45631 is a reply to message #45626] Thu, 17 December 2015 18:16 Go to previous messageGo to next message
Klugier is currently offline  Klugier
Messages: 1075
Registered: September 2012
Location: Poland, Kraków
Senior Contributor
Hello,

- First of all, one of the easiest channel to let user know about Ultimate++ is wikipedia. Currently the article is Čeština, Deutsch, English, Español, Nederlands, Português, Русский & Polski). A lot of this article is outdated.or a little bit too small. So my proposition is to create articles for other languages (community support is required) and update/upgrade current.Even small articles especially in Chinless, Indian and Japanese (huge countries with a lot of software potential).

- One of the problem is the heterogeneity of the brand. We have got too many brands that is hard to remember. We have Ultimate++, U++, TheIDE etc.. Especially the last brand (TheIDE) may cause misunderstandings. Maybe it should be something else that shows it is related to Ultimate++ like U++IDE Ultimate++ IDE etc..

- In 2016 Ultimate++ should participate in Google Summer of Code. Conclusions regarding past failures must be drawn. Google Summer of Code is good opportunity to attract new and perspective (young) developers.

- Backing to university, Android Builder (introduced in 2015.2 version) was my master thesis Wink (So it is definitely a possibility for the future - if you want to write thesis within Ultimate++ you can do this)

Sincerely,
Klugier


U++ - one framework to rule them all.

[Updated on: Thu, 17 December 2015 18:21]

Report message to a moderator

Re: What next? [message #45634 is a reply to message #45628] Thu, 17 December 2015 18:53 Go to previous messageGo to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
mr_ped wrote on Thu, 17 December 2015 17:56
Maybe... (just thinking loud) one possibility may be to pick up some popular unmaintained open source project, and write alternative replacement in U++ from scratch. But I have no idea about what project that could be, maybe some front end cross platform client for some open messaging API, like jabber/gtalk, including video calls, etc? That may have impact over larger user base, but looks like lot of work just to get it going, and without heavy maintenance it would probably die soon after.


Funny, just stumbled upon this:

https://sourceforge.net/p/keepass/discussion/329220/thread/1 bd4b047/?limit=25

(look at weekly downloads...)
Re: What next? [message #45635 is a reply to message #45631] Thu, 17 December 2015 19:07 Go to previous messageGo to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
Klugier wrote on Thu, 17 December 2015 18:16
Hello,

- First of all, one of the easiest channel to let user know about Ultimate++ is wikipedia. Currently the article is Čeština, Deutsch, English, Español, Nederlands, Português, Русский & Polski). A lot of this article is outdated.or a little bit too small. So my proposition is to create articles for other languages (community support is required) and update/upgrade current.Even small articles especially in Chinless, Indian and Japanese (huge countries with a lot of software potential).


I welcome anybody doing that. However, it is considered "bad" to use Wikipedia to promote own ideas, so I should not meddle with the article.

BTW, a couple of days ago, I have stumbled upon a very nice proposal for wikipedia article but some "wiki intern" or something. What a shame it was not published... (unfortunately, I fail to find it again).

Quote:

- One of the problem is the heterogeneity of the brand. We have got too many brands that is hard to remember. We have Ultimate++, U++, TheIDE etc.. Especially the last brand (TheIDE) may cause misunderstandings. Maybe it should be something else that shows it is related to Ultimate++ like U++IDE Ultimate++ IDE etc..


Actually, the name was likely a one of mistakes. It sounds like ranting. That is why later I have rather tried to promote U++.

In fact, maybe we should change the name.

Quote:

- In 2016 Ultimate++ should participate in Google Summer of Code. Conclusions regarding past failures must be drawn. Google Summer of Code is good opportunity to attract new and perspective (young) developers.


Maybe. But it would be just pointless PR stunt with little impact (who reads all these applications outside Google?). And Google will not choose us, because we are unknown.

Mirek
Re: What next? [message #45636 is a reply to message #45622] Thu, 17 December 2015 21:08 Go to previous messageGo to next message
dolik.rce is currently offline  dolik.rce
Messages: 1789
Registered: August 2008
Location: Czech Republic
Ultimate Contributor

mirek wrote on Thu, 17 December 2015 15:54
- Maybe release some parts of U++ as separate libraries. E.g. memory allocator is a good candidate IMO.

That is actually very good idea. I had a similar one few months ago, but never got time to actually try it. There is actually quite a lot of projects that use jemalloc or similar preloadable libraries to squeeze out some more performance out of their code. I believe "umalloc" could offer similar performance boost and it might help spread the word about the rest of U++ too.

Additionally, I think it might be worth to spend some time making U++ less dependent on TheIDE. Even though it is a great editor, many programmers are not willing to switch from their favorite IDE/editor (which they spent years learning and customizing) just to try a new framework. And using U++ without TheIDE is currently non trivial.

I remember some experiments with converting upp files to Visual Studio projects, waf based builds and of course the universal makefile, but frankly, all of those are quite difficult to use without previous knowledge of U++. Making a library that could be just linked against would help a lot, but I know this has already been tried several times before, with mixed results.

Best regards,
Honza
Re: What next? [message #45637 is a reply to message #45636] Fri, 18 December 2015 04:53 Go to previous messageGo to next message
Novo is currently offline  Novo
Messages: 1358
Registered: December 2006
Ultimate Contributor
IMHO, it would be a good idea to present about U++ at Meeting C++, CppCon, or CppNow. These events are very popular and a lot of people watch videos from these events and read slides. If you are presenting you do not need to pay fee. If Mirek cannot go there himself maybe somebody else can make a presentation about U++. meetingcpp.com has a list of C++ groups in different countries.

Regards,
Novo
Re: What next? [message #45638 is a reply to message #45622] Fri, 18 December 2015 05:19 Go to previous messageGo to next message
Novo is currently offline  Novo
Messages: 1358
Registered: December 2006
Ultimate Contributor
Two main competitors of U++ are Qt and wxWidgets. Both support Mac and can be built with all popular build systems.

Another thing. It is almost impossible to find any reference to U++ anywhere on Internet. It is just not listed among GUI and C++ frameworks.


Regards,
Novo

[Updated on: Sun, 20 December 2015 15:57]

Report message to a moderator

Re: What next? [message #45640 is a reply to message #45634] Fri, 18 December 2015 09:49 Go to previous messageGo to next message
mr_ped is currently offline  mr_ped
Messages: 825
Registered: November 2005
Location: Czech Republic - Praha
Experienced Contributor
mirek wrote on Thu, 17 December 2015 18:53
Funny, just stumbled upon this:

https://sourceforge.net/p/keepass/discussion/329220/thread/1 bd4b047/?limit=25


Interesting... now I recall I was recently (1m back?) googling a bit different terms around U++ (out of curiosity, to see the impact), already almost forgot about it.

The main line is, that it is extremely difficult to find anything... the term "ultimate" is so common, that if I would be first time TheIDE user, somehow ignorant of official web presence (like maybe just trying it out from repository, and not paying attention to "Help/..."), I would be unable to find anything.

Anyway, digging deeper into results I found few articles even outside this website, one of them is skype plugin, which went from ultimate++ in v0.20 (and using U++ only for linux version probably) to wxWidgets in v1.00:
http://clownfish-translator.com/linux.html


--
The thing about CppCon and similar sounds good to me, I sometimes check the videos too (especially Stroustrup's Smile)
Re: What next? [message #45642 is a reply to message #45638] Fri, 18 December 2015 10:59 Go to previous messageGo to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
Novo wrote on Fri, 18 December 2015 05:19
Two main competitors of U++ are Qt and wxWidgets. Both support Mac and can be built with all popular build systems.
There is also a new generation of GUI frameworks rising rapidly.

Another thing. It is almost impossible to find any reference to U++ anywhere on Internet. It is just not listed among GUI and C++ frameworks.


Hm, I thought this one is quite covered. Where is it missing?

Mirek
Re: What next? [message #45643 is a reply to message #45636] Fri, 18 December 2015 11:01 Go to previous messageGo to next message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
dolik.rce wrote on Thu, 17 December 2015 21:08

I remember some experiments with converting upp files to Visual Studio projects, waf based builds and of course the universal makefile, but frankly, all of those are quite difficult to use without previous knowledge of U++. Making a library that could be just linked against would help a lot, but I know this has already been tried several times before, with mixed results.


Well, I am afraid this is still mostly the question of INITBLOCK.

Should we drop INITBLOCK and require manual registration of modules used?

Mirek
Re: What next? [message #45652 is a reply to message #45642] Sat, 19 December 2015 04:37 Go to previous messageGo to next message
Novo is currently offline  Novo
Messages: 1358
Registered: December 2006
Ultimate Contributor
mirek wrote on Fri, 18 December 2015 04:59


Hm, I thought this one is quite covered. Where is it missing?

Mirek


Sorry, I'm wrong. U++ is listed on wikipedia. The rest is not important Smile


Regards,
Novo
Re: What next? [message #45657 is a reply to message #45622] Sat, 19 December 2015 14:23 Go to previous messageGo to next message
deep is currently offline  deep
Messages: 263
Registered: July 2011
Location: Bangalore
Experienced Member
One of the idea to increase visibility of U++

I think can we create projects / apps using U++ on Github / Sourceforge.

When people see useful app on Source forge developed using U++ it will increase visibility of the project.

There are many projects with wxWidgets on source forge


Warm Regards

Deepak
Re: What next? [message #45658 is a reply to message #45652] Sat, 19 December 2015 14:40 Go to previous messageGo to next message
cbpporter is currently offline  cbpporter
Messages: 1401
Registered: September 2007
Ultimate Contributor
I saw this thread when it was created, but it took me a few days to gather my thoughts on the subject. Doubly so, since I'm almost in the same boat, but probably where U++ was 10+ years ago: I too am developing a solution for some perceived problems taking the form of a library (no GUI though), an endeavor mostly fueled by dissatisfaction with current popular solutions. I am still in the development phase, but I'm dreading the day I will actually have to promote the project.

Because promotion is very hard, especially in this domain. Really really hard. So much harder than writing code Smile, which is easy and fun and has deterministic outcome. And especially now, when C++ is no longer as common as it used to be. Everybody is using JS now and we can't even hire competent new people at my workplace. The "if you build it, they will come" approach does not work. Consistently outputting a quality product does not attract people. It may be great at retaining people, but the first exposure most come from somewhere else.

U++ is a great tool with great productivity and generally a good design, but is is also often weird, quirky and some parts of it are downright alien. If you ask me, U++ is equally well designed and well designed to drive off random enthusiasts who just heard of the project and want to give it a quick try. The barrier of entry is too high.

I think that this is mostly due to TheIDE and a lack of structure that puts some of the best parts of U++ to the forefront.

U++ has great value as a GUI framework. But GUI work is relatively rare in this day and age. But it also has huge amounts of value outside of GUI. I love CodeEditor as an example, because it saved me from ritting thousand of lines of code, but you won't attract new people with CodeEditor or ArrayCtrl. You'll do it with Core at first on the non GUI side and with simple GUI samples that show of accessibility. GUI can be complicated and downright threatening looking in some toolkits, but creating a TopWindow (which should be called Window anyway) and plopping down a few Buttons can be inviting.

But I still think Core needs special attention. Core is super valuable but quite inaccessible.

Here is an example: I don't think that there is a seasoned C++ developer out there who really loves std::vector. Most people that I know hate it and iterators with pointer syntax in general. U++ has Vector, which is a lot more pleasant to use and potentially faster. I do not understand why after so many years, a random person can't just pick a random C++ source file, add a #include <upp/Vector> or something, and happily go on and use Vector. No mess, no fuss, no dependencies, no pulling in full packages, no random allocator, no almost proprietary IDEs. Or String. Or VectorMap. People do need hash maps all the time. std::map is so lacking. Or CParser. CParser is pure gold! I don't like pointless hyperbole, but CParser did change a small part of my life for the better! It is ridiculously convenient to have this power in a few lines of code without having to use some lexer lib and grammar. In a language notorious for having zero parsing ability without blowing up some pointers and barely being able to convert an int to a string. Or Socket. Very useful!

The contents of U++ should be filtered and a "core core" should be made very accessible. Not everything that is in the packages though. Just the super useful high quality stuff. Color is nice and I guess it does the job, but it is not a selling point. My Color lib is 40k code of code without conversion functions and I still find it lacking, so a simple Color class is not an attraction like the containers, serialization, Xml support, Unicode and so on.

Now I'm not saying that we no longer use packages or TheIDE. I'm saying the message for new initiates should be "U++ Core is awesome because these classes are great. You can use TheIDE for a modular approach at developing programs, but first let me show you how awesome Vector is using the Linux command line/CMake/Visual Studio. You'll love it and never touch std::vector again! And BTW, Vector is compatible with std, so you can still mix it with other code!".

Neither am I saying we break up Core into 2 separately maintained code bases. There is probably a way to reorganize sources so that stand alone classes can be included with standard C++ paradigms (some thought must be put into how to break up libs for non templates, probably the boost model) one by one and still have Core.h which resides in a U++ package and provides the entire functionality.

[Updated on: Sat, 19 December 2015 14:42]

Report message to a moderator

Re: What next? [message #45661 is a reply to message #45622] Sun, 20 December 2015 00:31 Go to previous messageGo to next message
Sgifan is currently offline  Sgifan
Messages: 37
Registered: February 2010
Location: France
Member
Communicate about the good things about UPP:

Productivity gains are huge for a C++ developer. The large scope of the framework really helps also in this area.
Being able to use a unique framework for serial communication, databases handling, charts and reports for instance
has allowed me to gain a great deal of time.

Try to enhance some not so good points:

TheIDE is great but could try to be a bit more simple concerning navigation (finding references, definitions, usages, etc...) (I quite like NetBeans these days in this area).
I would like to see theIDE be capable of having multiple targets defined inside a project, and one of these target could be a unit tests.
Also having unit tests directly supported in theIDE gui.


The web site is a bit old fashioned and I found Github so nice I really think there is good inspiration to take from there.
Having a more dynamic web site to be able to use it from a mobile device would be great.

The advertising on the site is also a bit strange (I get for instance a lot of dating advertising on the left side of the screen)

[Updated on: Sun, 20 December 2015 00:33]

Report message to a moderator

Re: What next? [message #45662 is a reply to message #45661] Sun, 20 December 2015 10:00 Go to previous messageGo to next message
Mindtraveller is currently offline  Mindtraveller
Messages: 917
Registered: August 2007
Location: Russia, Moscow rgn.
Experienced Contributor

I guess it is time we (U++ users) should help Mirek.

1. First of all, we must update comparison on U++ site adding more comparisons showing more cases of U++ usage vs Qt/wxWidgets. Comparisons must be 'honest' even if U++ shows little/no advantage in the particular scenario.

2. Second, we must start writing 'U++ red book' showing methodology and techniques of different aspects of U++ programming. Especially those which are very different from widely used ones.
This is something very different from U++ help. It is more a teaching book than a reference.
I believe this is something U++ needs a lot. U++ is very hard to learn mostly because the most of knowledge is scattered among this forum and is hard to find.

3. Each U++ user should write 1-2-3 or more articles on different interesting aspects of U++, preferably one in it's own language and the second one in English. Topics include: new U++ features (for the recent year), Rainbow/Mac/Android compatibility, Core containers, etc.

This is really something we (users) may do without passing it on to Mirek.
Re: What next? [message #45663 is a reply to message #45631] Sun, 20 December 2015 10:08 Go to previous messageGo to next message
Mindtraveller is currently offline  Mindtraveller
Messages: 917
Registered: August 2007
Location: Russia, Moscow rgn.
Experienced Contributor

Klugier wrote on Thu, 17 December 2015 21:16
- One of the problem is the heterogeneity of the brand. We have got too many brands that is hard to remember. We have Ultimate++, U++, TheIDE etc.. Especially the last brand (TheIDE) may cause misunderstandings. Maybe it should be something else that shows it is related to Ultimate++ like U++IDE Ultimate++ IDE etc..


I agree. My proposal is that we concentrate on 'U++'. U++ is a framework. 'U++ ide' for the ide.
I doubt if calling 'Ultimate' is a good approach but it IMO it is completely up to U++ authors to decide.
Frankly speaking I guess it is not very important at all.
Re: What next? [message #45664 is a reply to message #45663] Sun, 20 December 2015 10:46 Go to previous messageGo to previous message
mirek is currently offline  mirek
Messages: 13975
Registered: November 2005
Ultimate Member
Thanks everybody for posting ideas. I will try to summarize now, and also try to estimate 'costs'

PR and Docs

- try to publish articles on codeproject or else. Cost: very low
- explain Core, compare with standard library. Cost: medium
- write Red Book. Might be the same as previous. Cost: medium
- fix names/brands. Cost: low to medium. Moderate risk involved.
- update wikipedia article. Cost: low. Low risk involved
- update website. Cost: high
- more comparisons, benchmarks. Cost: low


Public events/people

- universities. Cost: unknown.
- C++ meetings. Cost: low (local) to high (intercontinental)


Programming "ex" U++

* create libraries based on U++
- just allocator. Cost: low
- Core subset. Cost: medium
- whole packages. Cost: high
- promote U++ by creating public projects in it: Cost: very high


Programming U++ (this, however, is sort of "business as usual" Smile

- support MacOS: Cost: very high, but it should definitely happen
- support Android: Cost: very high, but it should definitely happen
* improve theide
- support more targets. Cost: medium
- further improve Assist++ (to support references etc.). Cost: very high (but most fun Smile

[Updated on: Sun, 20 December 2015 10:49]

Report message to a moderator

Previous Topic: Members of structure are finally accessible through plain index
Next Topic: I've just joined the wonderful world of Linux software distribution!
Goto Forum:
  


Current Time: Thu Mar 28 10:41:48 CET 2024

Total time taken to generate the page: 0.01109 seconds