Alright.Thanks for replying.I can't help about those problems you mentioned; I'm looking ahead to develop new software and/or services, that's what makes me excited.

I have to ponder a bit more about this whole KWE thing and talk to my countrymen at the international forum, may I get an idea where to start off.

All I know is that I'm learning JSP/JSF and Java on the whole and I need to put my knowledge -however deep it is- to work somehow .You know, I'd like to pursue my idea.

Also about your comment on C++ being easier that Java , I've heard and read that Java has a much steeper learning curve than that of C++.Even one of the stated goals of Sun was to create a higher level version of C++.

Madman wrote:If the way of the future is the web browser, then it's a bleak future indeed...

Try to use a good browser with Google Gears support and, maybe, you can change your mind. If not, don't worry, even mouse or graphics desktops have detractors in the past.

When in KOffice, or other spreadsheet application, three people in two different places of the country could update the same spreadsheet concurrently, and your sales man can seeing the changes in real-time using a mobile, then I agree with you. Meanwhile, I see a shining future in web applications and a slow decline in desktop applications.

Ignacio Serantes wrote:When in KOffice, or other spreadsheet application, three people in two different places of the country could update the same spreadsheet concurrently, and your sales man can seeing the changes in real-time using a mobile, then I agree with you. Meanwhile, I see a shining future in web applications and a slow decline in desktop applications.

Now, THAT's the kind of future I'm looking forward to.
It doesn't even matter which browser you use, though: Google Docs is, in and of itself, very primitive. There are just so many ways that KOffice is nicer then Google Docs: images with text-align and on-the-fly updating of position, rotating/scaling images, basic drawing tools (which is useful for illustration), the ability to copy-paste any old shape from one type of document into another (spreadsheets/charts into a word document) and have them behave in a unified, standard manner (you can resize/rotate/skew any shape, regardless of content)...

And in all honesty, I can never see Google Docs become better then a fully dedicated office suite like KOffice - the only real benefit is collaborative editing, which is slowly coming to dedicated desktop applications anyway (OOo calc, Abiword). KOffice with collaborative editing is my idea of, "the future of office suites".

Madman wrote:And in all honesty, I can never see Google Docs become better then a fully dedicated office suite like KOffice - the only real benefit is collaborative editing, which is slowly coming to dedicated desktop applications anyway (OOo calc, Abiword). KOffice with collaborative editing is my idea of, "the future of office suites".

The same objective but with opposite points of view. The problem with your point of view is that you need a powerful infrastructure to avoid KOffice application works, in our example you need KDE working over a heavy OS, and with my point of view you only need a browser working over any kind of OS. And you're forgetting that a server is needed so, witch technology will you use to share data?, low level tcp/ip connection? web services?, a db server? or you are planning to use p2p data transmission?.

Actually all kind of new hardware (computer, mobile, video consoles, ...) have a browser because you need few resources to launch one (I hear that there is BIOS with a small linux and a firefox out there) so all this hardware can work with web applications without installing anything. In my case, I use my NDS sometimes to read my mail or my feeds so I don't use Kontact to do this things because I don't live chained to my PC.

Obviously both points of view have it's flaws, but, in our particular case our priorities are collaborative applications that can be used in any kind of OS or hardware so, if we lose power because this we don't care. Naturally, for a person working isolated from the world in his PC with his OS and his favorite application, power is more important than sharing.

Ignacio Serantes wrote:

Madman wrote:And in all honesty, I can never see Google Docs become better then a fully dedicated office suite like KOffice - the only real benefit is collaborative editing, which is slowly coming to dedicated desktop applications anyway (OOo calc, Abiword). KOffice with collaborative editing is my idea of, "the future of office suites".

The same objective but with opposite points of view. The problem with your point of view is that you need a powerful infrastructure to avoid KOffice application works, in our example you need KDE working over a heavy OS, and with my point of view you only need a browser working over any kind of OS. And you're forgetting that a server is needed so, witch technology will you use to share data?, low level tcp/ip connection? web services?, a db server? or you are planning to use p2p data transmission?.

To start: KDE is actually incredibly flexible. It is very capable of being a light-weight DE. What's inflexible is the package management systems of several distributions, which require that you install ALL libraries for ALL applications in a meta-group like kdebase. The differences between Arch linux and something like Kubuntu illustrate this quite well, I think (Arch is reportedly capable of a light-weight KDE desktop, since it allows you to have a very minimalistic desktop and build on top of it, excluding what you don't need). Besides, the future I envision is inter-operability between several office suites, such as collaborative editing becoming part of the ODT spec, so that you can choose between office suites depending on your machine, requirements and taste.
Besides, I'd use something like Dropbox/Ubuntu One for document collaboration, myself.

Ignacio Serantes wrote:Actually all kind of new hardware (computer, mobile, video consoles, ...) have a browser because you need few resources to launch one (I hear that there is BIOS with a small linux and a firefox out there) so all this hardware can work with web applications without installing anything. In my case, I use my NDS sometimes to read my mail or my feeds so I don't use Kontact to do this things because I don't live chained to my PC.

Obviously both points of view have it's flaws, but, in our particular case our priorities are collaborative applications that can be used in any kind of OS or hardware so, if we lose power because this we don't care. Naturally, for a person working isolated from the world in his PC with his OS and his favorite application, power is more important than sharing.

Madman wrote:To start: KDE is actually incredibly flexible. It is very capable of being a light-weight DE. What's inflexible is the package management systems of several distributions, which require that you install ALL libraries for ALL applications in a meta-group like kdebase. The differences between Arch linux and something like Kubuntu illustrate this quite well, I think (Arch is reportedly capable of a light-weight KDE desktop, since it allows you to have a very minimalistic desktop and build on top of it, excluding what you don't need). Besides, the future I envision is inter-operability between several office suites, such as collaborative editing becoming part of the ODT spec, so that you can choose between office suites depending on your machine, requirements and taste.
Besides, I'd use something like Dropbox/Ubuntu One for document collaboration, myself.

First of all, KDE don't works, and probably never do, in my Wii, my mobile, my NDS, my PS3, etc..., so, well, your application will have an important target limitation. You still have a PC centric vision and, well, I think is time to evolve and think a few in a new generation of devices capable to doing things. I'm not sure if I right or not but, I think that any project that ignore this devices probably will fail in medium/long term.

Second, you obviously don't understand anything about Google Apps posibilities. We use Dropbox so I know a lot about it's problems because you must manually handle update conflicts. The same problem exist working with a file in a file server if this file is not locked by the server. With Google spreadsheet more than one people can work concurrently.

Honestly, I don't see any advantage in your proposition because don't offers anything new and have the same problems actually exists with desktop applications.

The advantage in my position is choice. If someone makes a doc in google docs and wants other people to collaborate, they all need google accounts to do it. You can't set up your own server for something similar, you can't work with your preferred/a familiar office suite to edit the document.

Besides, Plasma was designed with scalability in mind - the Nokia N900 is one of the first small-factor platforms it's been ported to, due to the debian base - and not only does it work, it works incredibly well. I expect that Plasma, as well as the other KDE components, will be ported to just about anything supported by Qt - including small/embedded devices. Moreover, KOffice on the N900 is fast, lightweight and efficient, showing that KDE applications like KOffice can be easily scaled to the small form-factors.

And as far as I know: the Nintendo Wii doesn't come with a web browser (at least, the one I got didn't seem to - I never found one in all the time I used it); the PS3 allowed you to use Linux on it anyway (giving you any application available on Linux, including KOffice); and the NDS doesn't include a web browser either.

Sure, a browser-based office suite might suit small/obscure form factors - but it is just so much nicer to be using a full, dedicated application.

Madman wrote:The advantage in my position is choice. If someone makes a doc in google docs and wants other people to collaborate, they all need google accounts to do it. You can't set up your own server for something similar, you can't work with your preferred/a familiar office suite to edit the document.

That's true if you can't download your document from Google Apps and open with an application of your choice, do the changes and import it again. But I think you misunderstood me, I put Google Apps as example because was the system we use. Other companies has other solutions. Talk is about web application and not Google Applications.
Obviously you need an account in the server but, in your case, you need to install kde+koffice+dropbox over linux. In one side I need an account and a browser, on the other side I need a desktop environment, an application and, finally, a (limited) sharing data utility. Can you see the difference?
At last, I assumed that you are talking about KOffice 1 because KOffice 2 was..., not ready.

Madman wrote:Besides, Plasma was designed with scalability in mind - the Nokia N900 is one of the first small-factor platforms it's been ported to, due to the debian base - and not only does it work, it works incredibly well. I expect that Plasma, as well as the other KDE components, will be ported to just about anything supported by Qt - including small/embedded devices. Moreover, KOffice on the N900 is fast, lightweight and efficient, showing that KDE applications like KOffice can be easily scaled to the small form-factors.

A particular case and, well, if this applications don't works in Nokia mobiles it's time to throw in the towel. Unfortunately not all people in the planet will have a Nokia and I'm not sure about future evolution of Nokia. Time pass and I can't see a success in sales alternative to iPhone or Android.
You mention scalability and I wonder why stability was not in mind when Plasma was designed. With current design, the only solution to make Plasma stable is limit what and how many plasmoids is running.

Madman wrote:And as far as I know: the Nintendo Wii doesn't come with a web browser (at least, the one I got didn't seem to - I never found one in all the time I used it); the PS3 allowed you to use Linux on it anyway (giving you any application available on Linux, including KOffice); and the NDS doesn't include a web browser either.

Both Wii and NDS has web browsers, an Opera browser in both cases, check this was really easy using google, on the other side, the fact that you mention install linux on PS3 shows that you really have a big PC centric perspective of computing. Future of applications will be in servers and not in clients.

Madman wrote:Sure, a browser-based office suite might suit small/obscure form factors - but it is just so much nicer to be using a full, dedicated application.

Sure, some people need a scientific calculator with graphics to work but, use (or buy) a tool like these to do the four basic operations is not wise. Several desktop applications are fat, heavy and with thousand of options that only few people use. Google applications has few options but, you can write a text, format it and even print it, and the same with spreadsheets with graphics. Enough to us.

And, at last, my friend, "nicer to use" is a relative term so, to defend your position is better that you use non relative arguments. Only Einsten could be relative and to be right .

Ignacio Serantes wrote:That's true if you can't download your document from Google Apps and open with an application of your choice, do the changes and import it again. But I think you misunderstood me, I put Google Apps as example because was the system we use. Other companies has other solutions. Talk is about web application and not Google Applications.
Obviously you need an account in the server but, in your case, you need to install kde+koffice+dropbox over linux. In one side I need an account and a browser, on the other side I need a desktop environment, an application and, finally, a (limited) sharing data utility. Can you see the difference?
At last, I assumed that you are talking about KOffice 1 because KOffice 2 was..., not ready

This is growing increasingly irritating, because YOU are missing MY point.
MY point is that you wouldn't NEED KOffice, KDE and Linux to collaborate. My point is that you would be able to collaborate with KOffice, KDE and Linux if you so choose, but if you didn't want to you would be able to use MSOffice on Windows, connect to the same server, open the same files and collaborate live in exactly the same way. Or, you could use OpenOffice.org on Gnome in FreeBSD, connect to the server in the same way, open the same document and collaborate in the same way. Or, you could use KOffice on the N900, connect to the server, open the document and collaborate in the same way. This would provide the power, features and integration of desktop applications as well as the benefits of collaborative editing: for example, Nepomuk may well be able to tell who has edited the document if I'm using such a system in KOffice, but it may be harder to implement that into something like Google Docs.

So, to summarize the two sides:
Madman's position - online server (managing the collaboration), client side applications (editing of documents). There would be a public api for connecting to the server and transferring of changes between the server and the clients. The public api would be implemented by whatever office suite i.e. KOffice, OOffice, MSOffice (even Google docs or other online office suite potentially, but Madman didn't explicitly mention this as a possibility). Pros: use whatever office suite you want (that implements the api), KOffice 2 (in the future when it has reached feature-completion) will be available on anything that runs QT. Cons: many devices don't run and won't likely have built-in office suites.

Ignacio Serantes's position - online server, server-side clients interfaced via web browser. Pros: usable with anything which runs a browser (read- more cross-platform than QT). Cons: current examples of online-only office suites are rather limited, esp. as compared to prevailing desktop-only office suites, managing a dedicated online office application suite is a pain (read- server maintenance, online security, browser interoperability, etc.).

How about we marry the two ideas? Have an online server for managing the collaboration which has a public api. Implement the public api in both the desktop office suites (K/O/MSOffice) AND an online office suite (KWO - KDE Web Office). KOffice will be designed to be able to run on as many platforms as possible and provide comprehensive functionality, while KWO will provide browser-based capabilities to be available where KOffice is not. KWO should not be expected to be feature-complete, but provide sufficient functionality to allow whatever device has a browser to be able at least view documents (and collaborate if possible).

airdrik wrote:How about we marry the two ideas? Have an online server for managing the collaboration which has a public api. Implement the public api in both the desktop office suites (K/O/MSOffice) AND an online office suite (KWO - KDE Web Office). KOffice will be designed to be able to run on as many platforms as possible and provide comprehensive functionality, while KWO will provide browser-based capabilities to be available where KOffice is not. KWO should not be expected to be feature-complete, but provide sufficient functionality to allow whatever device has a browser to be able at least view documents (and collaborate if possible).

I'm not against web-based applications per se, but I'm against making the web-browser THE standard application for accessing such on-line services, yes.

Madman wrote:This is growing increasingly irritating, because YOU are missing MY point.
MY point is that you wouldn't NEED KOffice, KDE and Linux to collaborate. My point is that you would be able to collaborate with KOffice, KDE and Linux if you so choose, but if you didn't want to you would be able to use MSOffice on Windows, connect to the same server, open the same files and collaborate live in exactly the same way. Or, you could use OpenOffice.org on Gnome in FreeBSD, connect to the server in the same way, open the same document and collaborate in the same way. Or, you could use KOffice on the N900, connect to the server, open the document and collaborate in the same way. This would provide the power, features and integration of desktop applications as well as the benefits of collaborative editing: for example, Nepomuk may well be able to tell who has edited the document if I'm using such a system in KOffice, but it may be harder to implement that into something like Google Docs.

Sorry but I don't try to irritate you. This was a talk and I think that your point is utopic and I disagree to your darkness vision of future with web applications.

This year, i was in a KOffice presentation with a KWord 2 developer and he exposes how difficult was to be compatible even with OpenOffice odf in a mono user environment. An api to avoid work with all this applications you mention that offer the same result and work with different application technology designs sounds, for me, impossible to avoid. Never say never but I can't see working that environment in real-time with concurrency, because is too difficult to achieve.

About Nepomuk, well, I have nepomuk in google apps now, or something similar. I can tag my mails, my documents, my spreadsheets and I can do a search and share with my workgroup. Nepomuk in KDE 4.3.2 has big problems with strigi and search is in early stages. There is to much work to do with nepomuk to be usable in a production environment. By the way, sharing is one of the big flaws of nepomuk implementation because I can't share my nepomuk data with other users in the same server.

On the other side, you are forgetting RIA applications, developed using Flesh or Silverlight. This technologies are relative new and they don't be mature but, I think that soon there will be an explosion of applications like these. On the other side, Zimbra shows that complex web interfaces could be implemented with ajax and, with Chrome or Firefox, the browsers I use, performance was accurate.

airdrik wrote:How about we marry the two ideas? Have an online server for managing the collaboration which has a public api. Implement the public api in both the desktop office suites (K/O/MSOffice) AND an online office suite (KWO - KDE Web Office). KOffice will be designed to be able to run on as many platforms as possible and provide comprehensive functionality, while KWO will provide browser-based capabilities to be available where KOffice is not. KWO should not be expected to be feature-complete, but provide sufficient functionality to allow whatever device has a browser to be able at least view documents (and collaborate if possible).

I explain in my previous post, this solution is really difficult to achieve because a really complex api was needed and all must be develop and standard implementation. I think that is not necessary to mention here problems associated with this kind of stuff. Javascript, DOM or HTML was good examples of these.

But I don't try to be radical, obviously, web application was not the perfect solution to all environments and, in that cases, desktop application probably would be the solution. The point is how thing would be in the future and I think that is easy to determine expending a few moment seeing our children. Studying how many ours spend with his game console, mobile or PC and you can see how many time spend with a PC when he grown.

All have his preferences and mine is that all my applications would be available at anytime, in anyplace with any hardware.