I'm also a consultant and I know that professionally, number of platforms is a really big Linux issue.
Right, so as a consultant, *you* do the choosing for your clients. It is up to *you* to choose and recommend the solution that matches the needs of your client. This is the job of consultants in every industry, and in every industry there are almost always competing solutions to any particular problem.
So pick one
For the open source community as a whole, that is never going to happen. It's not an open source thing, it's a human nature thing. Of course, within the community there will be smaller groups that make choices, eg. Ubuntu choosing GNOME as its desktop environment - in fact most distributions make a choice of the default and/or supported desktop.
picking standards and increasing interoperability is a very big part of the effort.
Absolutely, and this is already happening. In quite a few key areas the two desktop platforms are already cooperating on standards and other areas of common ground; but it is unrealistic for you to expect one camp to throw away everything and basically say "whoops, sorry everyone - we got it completely wrong, the other camp were right so we'll use all their stuff now.". Of course that's an exaggeration, but to me that's pretty much what you'd like to see happen.
It seems to me that this is more of a marketing problem. Perhaps if you, the consultant, were to push "the KDE desktop" or "the GNOME desktop", or heck even "the Ubuntu desktop" instead of "the Linux desktop", the issue of competing solutions would not even need to be brought up with your clients.
this is the exact problem, whoever made Kubuntu does not want Linux to succeed as the consumer desktop environment of choice.
So basically if a distribution such as Ubuntu makes a decision about the right direction to go in, I as a developer should just automatically follow them even if what they are doing conflicts with what I believe is the correct decision? Right.
What you are suggesting not only goes against open markets (where companies can have competing products and competition drives innovation), it goes against human nature itself - one person's idea of the "One True Way"(TM) is not necessarily going to be the same as another's.
One thing that a lot of people don't seem to understand when they say that "KDE and GNOME should just merge" or even "why are there two desktop platforms for Linux?" is that they are not just two products providing the same or similar feature set. For developers (as well as systems integrators and distributors) choosing one over the other implies choosing quite a different path in terms of programming language, API, design philosophy, tools etc. For those people it's not necessarily going to just be about what it looks like or how it feels, it's what it's like to develop for and its roadmap for future development.
Regarding IPv6, the cynic in me says that the actual reasoning is along similar lines: IP addresses will no longer be a scarce resource and therefore providers won't be able to charge as much for static ones - so why would they spend all the money to implement IPv6 when it'll probably lose them money anyway?
From his wikipedia entry: With a long background in open source, he has worked to build bridges between Microsoft and the Linux and Open Source communities both politically and also technically.
I can see it now, billboards with signs saying: "a microwave is a great way to get a tan," "a blowtorch is a great way to thaw frozen pipes," and of course "it is cool to operate power tools--in the nude! Buy Budweiser!"
Don't forget "did you know: used motor oil can be used to fertilize your lawn".
The Linux and Mac Programmers haven't written one yet.
And without an open (and easily implementable, to counter OpenXML) specification, it's very hard for them to do that. This is the whole problem with proprietary formats.
You'll need to provide more details, because I just tried exactly what you said and it worked perfectly. I created two columns with dummy data, selected the data, clicked on add chart, selected the area for the chart to go and then selected the line graph style. Perhaps you are doing something different?
PLEASE start using a dialog manager that arranges the "ok" and "cancel" buttons according to a single policy?
In Qt 4.x Trolltech has fixed Qt to use the appropriate button order when running within GNOME. Unfortunately I haven't read anything about Gtk doing the same when running in KDE.
CUPS is pretty horrifying, but for the worst of the "edit an obscure config file to make basic stuff work" nightmare of yesteryear, look no further than ALSA, which needs to be taken out behind the barn and shot. Repeatedly.
After thrashing about for several hours last night trying to get surround sound to work on my new sound card, I couldn't agree more. ALSA is very powerful, but there's no GUI interface for it that I've seen that sets everything up properly for you *and* lets you easily do fancy stuff like remapping speaker channels. This is what is really needed I think.
applications not showing on iconification when also shaded
Interesting, I had never noticed that - but to be honest, shading is a feature that I have never thought was particularly useful. In your parents' case I would suggest changing the double-click behaviour so that they never use the shading by accident (which is the only way I ever used to activate it) and then they'll never run into the problem.
To be fair, KDE has a much greater usability focus now than ever, and usability bugs do get fixed. Perhaps the particular ones you mention have just been forgotten - if nobody else has, perhaps you could add a comment mentioning that the bug is still present in which ever version you are using.
A lot of the things you mention are coming in KDE 4. In particular the system-wide searching (Kerry/Beagle, Strigi, etc.) and hardware / system integration through Solid, such as not doing silly stuff like power saving in the middle of a presentation, and network profiles.
1. Just click OK. It's not that difficult. If you close the dialog or click Cancel you're telling the application you aren't sure of your changes.
4. Most (if not all) KDE applications have Settings -> Customise Shortcuts. This will let you customise all of the shortcuts for the application. I admit, being able to do it from the context menu on a toolbar button would be nice though.
5. Well, for me the open/save dialogs are a hell of a lot better than those you find in GTK+ applications, because they're far more flexible.
6. For me, KDE's integration and consistency far outweigh any minor "problems". In all honesty though it sounds like the Gnome style of sparse options is more suited to your tastes, so you should probably stick with it (I'm assuming that's what your current DE is).
Just out of curiosity -- HAVE you contacted the developer asking for a fix? Just because its a closed-source solution you can't fix yourself, doesn't mean the vendor won't fix it if someone asks.
Unfortunately not all closed source vendors are as helpful to their customers as your company. I once dealt with a problem in a closed-source accounting package, which could not handle a fairly simple way of grouping items together to be sold (selling a specified set of items as a "kit" at a reduced price). I contacted the firm that developed the software and asked them if they were planning on adding this feature - no. Would they consider adding it for us for a fee - no. "Not for any amount of money?" I asked. "No." That was the end of the conversation.
If you're suggesting that wxWidgets even approaches the capabilities or ease of development of Qt, then you've probably never used Qt to develop software or seen what it can do. Just take a look at some of the new features in Qt 4.2 alone. Qt's strength is in its capabilities, quality, and the saving of development time. Trolltech's claim is that as a commercial developer using Qt you'll save more money than you'll spend on licensing, and I think it's not an unreasonable claim to make.
Well, turn it around then: if we received a signal from a distant alien race, don't you think we'd put our very best people onto decoding it the very same day it was received? You bet we would.
Your point about our safety is well made. However, what if early explorers decided it was too dangerous to go out? "Nah, there might be monsters in those hills, I think i'll stay home by the fire". Safe, but pretty dull when there might be interesting things to discover.
Slashdot Mirror
Well, I'd hope that privacy was sanctioned by governments, but with governments being what they are these days...
Wait a sec, did you mean *piracy*?
I'm also a consultant and I know that professionally, number of platforms is a really big Linux issue.
Right, so as a consultant, *you* do the choosing for your clients. It is up to *you* to choose and recommend the solution that matches the needs of your client. This is the job of consultants in every industry, and in every industry there are almost always competing solutions to any particular problem.
So pick one
For the open source community as a whole, that is never going to happen. It's not an open source thing, it's a human nature thing. Of course, within the community there will be smaller groups that make choices, eg. Ubuntu choosing GNOME as its desktop environment - in fact most distributions make a choice of the default and/or supported desktop.
picking standards and increasing interoperability is a very big part of the effort.
Absolutely, and this is already happening. In quite a few key areas the two desktop platforms are already cooperating on standards and other areas of common ground; but it is unrealistic for you to expect one camp to throw away everything and basically say "whoops, sorry everyone - we got it completely wrong, the other camp were right so we'll use all their stuff now.". Of course that's an exaggeration, but to me that's pretty much what you'd like to see happen.
It seems to me that this is more of a marketing problem. Perhaps if you, the consultant, were to push "the KDE desktop" or "the GNOME desktop", or heck even "the Ubuntu desktop" instead of "the Linux desktop", the issue of competing solutions would not even need to be brought up with your clients.
this is the exact problem, whoever made Kubuntu does not want Linux to succeed as the consumer desktop environment of choice.
So basically if a distribution such as Ubuntu makes a decision about the right direction to go in, I as a developer should just automatically follow them even if what they are doing conflicts with what I believe is the correct decision? Right.
What you are suggesting not only goes against open markets (where companies can have competing products and competition drives innovation), it goes against human nature itself - one person's idea of the "One True Way"(TM) is not necessarily going to be the same as another's.
One thing that a lot of people don't seem to understand when they say that "KDE and GNOME should just merge" or even "why are there two desktop platforms for Linux?" is that they are not just two products providing the same or similar feature set. For developers (as well as systems integrators and distributors) choosing one over the other implies choosing quite a different path in terms of programming language, API, design philosophy, tools etc. For those people it's not necessarily going to just be about what it looks like or how it feels, it's what it's like to develop for and its roadmap for future development.
Regarding IPv6, the cynic in me says that the actual reasoning is along similar lines: IP addresses will no longer be a scarce resource and therefore providers won't be able to charge as much for static ones - so why would they spend all the money to implement IPv6 when it'll probably lose them money anyway?
If you think "Joe Bob" has $3000 to spend on a digital camera, you're deluding yourself.
The Neo1973 does have GPS, both in the GTA-01 and GTA-02.
You know, I don't find this stuff amusing anymore.
From his wikipedia entry:
With a long background in open source, he has worked to build bridges between Microsoft and the Linux and Open Source communities both politically and also technically.
Way to build those bridges there, Bill.
But if anything to do with the argument to go with a particular operating system is "we'll save money on developers" then this is what happens.
Unnecessarily complicated system for the task + cheap programmers = nightmare. You can't argue with that.
I can see it now, billboards with signs saying: "a microwave is a great way to get a tan," "a blowtorch is a great way to thaw frozen pipes," and of course "it is cool to operate power tools--in the nude! Buy Budweiser!"
Don't forget "did you know: used motor oil can be used to fertilize your lawn".
The Linux and Mac Programmers haven't written one yet.
And without an open (and easily implementable, to counter OpenXML) specification, it's very hard for them to do that. This is the whole problem with proprietary formats.
You'll need to provide more details, because I just tried exactly what you said and it worked perfectly. I created two columns with dummy data, selected the data, clicked on add chart, selected the area for the chart to go and then selected the line graph style. Perhaps you are doing something different?
PLEASE start using a dialog manager that arranges the "ok" and "cancel" buttons according to a single policy?
In Qt 4.x Trolltech has fixed Qt to use the appropriate button order when running within GNOME. Unfortunately I haven't read anything about Gtk doing the same when running in KDE.
CUPS is pretty horrifying, but for the worst of the "edit an obscure config file to make basic stuff work" nightmare of yesteryear, look no further than ALSA, which needs to be taken out behind the barn and shot. Repeatedly.
After thrashing about for several hours last night trying to get surround sound to work on my new sound card, I couldn't agree more. ALSA is very powerful, but there's no GUI interface for it that I've seen that sets everything up properly for you *and* lets you easily do fancy stuff like remapping speaker channels. This is what is really needed I think.
So, you're obviously quite keen on a compulsory ID system. Why is that?
http://imdb.com/title/tt0238380
A truly brilliant film.
applications not showing on iconification when also shaded
Interesting, I had never noticed that - but to be honest, shading is a feature that I have never thought was particularly useful. In your parents' case I would suggest changing the double-click behaviour so that they never use the shading by accident (which is the only way I ever used to activate it) and then they'll never run into the problem.
To be fair, KDE has a much greater usability focus now than ever, and usability bugs do get fixed. Perhaps the particular ones you mention have just been forgotten - if nobody else has, perhaps you could add a comment mentioning that the bug is still present in which ever version you are using.
A lot of the things you mention are coming in KDE 4. In particular the system-wide searching (Kerry/Beagle, Strigi, etc.) and hardware / system integration through Solid, such as not doing silly stuff like power saving in the middle of a presentation, and network profiles.
I'll answer the ones I can answer:
1. Just click OK. It's not that difficult. If you close the dialog or click Cancel you're telling the application you aren't sure of your changes.
4. Most (if not all) KDE applications have Settings -> Customise Shortcuts. This will let you customise all of the shortcuts for the application. I admit, being able to do it from the context menu on a toolbar button would be nice though.
5. Well, for me the open/save dialogs are a hell of a lot better than those you find in GTK+ applications, because they're far more flexible.
6. For me, KDE's integration and consistency far outweigh any minor "problems". In all honesty though it sounds like the Gnome style of sparse options is more suited to your tastes, so you should probably stick with it (I'm assuming that's what your current DE is).
The default look satisfies me, so that's at least one. I'm sure it could be improved but it's not that bad at the moment.
Damn, you beat me to it.
Interestingly though the painting they show in the film could not have been the real one as it's too small.
Just out of curiosity -- HAVE you contacted the developer asking for a fix? Just because its a closed-source solution you can't fix yourself, doesn't mean the vendor won't fix it if someone asks.
Unfortunately not all closed source vendors are as helpful to their customers as your company. I once dealt with a problem in a closed-source accounting package, which could not handle a fairly simple way of grouping items together to be sold (selling a specified set of items as a "kit" at a reduced price). I contacted the firm that developed the software and asked them if they were planning on adding this feature - no. Would they consider adding it for us for a fee - no. "Not for any amount of money?" I asked. "No." That was the end of the conversation.
If you're suggesting that wxWidgets even approaches the capabilities or ease of development of Qt, then you've probably never used Qt to develop software or seen what it can do. Just take a look at some of the new features in Qt 4.2 alone. Qt's strength is in its capabilities, quality, and the saving of development time. Trolltech's claim is that as a commercial developer using Qt you'll save more money than you'll spend on licensing, and I think it's not an unreasonable claim to make.
... use a long spoon?
Well, turn it around then: if we received a signal from a distant alien race, don't you think we'd put our very best people onto decoding it the very same day it was received? You bet we would.
Your point about our safety is well made. However, what if early explorers decided it was too dangerous to go out? "Nah, there might be monsters in those hills, I think i'll stay home by the fire". Safe, but pretty dull when there might be interesting things to discover.
Even a journalist couldn't get an article about the "mistake" right...