The main reason why you run into multiple incompatible mechanisms in software like Emacs is that the software started as a TTY application,
Right, and if clipping was recognised as nothing to do with a GUI, just a bit of data storage, then TTY apps could use it.
It sounds to me as though you are starting multiple independent instances of Xemacs to edit multiple files - I think that you will find that running a single Emacs process and opening multiple "frames" to edit multiple buffers is the correct way to go.
I've seen those frames and don't like them much. If you don't have a windowing system, then I suppose that they would be a necessary substitute.
The basic problem is that different instances of the same program, let alone different ones, don't play nice together.
It CAN be used by any... that uses X... so what's the problem?
That. The mechanics of storing clips has got nothing to do with a gui of any kind, X or otherwise. It should be independant of X, ie usable from commandline vi and emacs.
Actually, the X has a fairly sophisticated clipboard
And that's where the problems start. Thee is nothing about storing smallish pieces of data temporarily that needs to be part of any gui or windowing toolkit - it's not a visual thing in any way. It should be a system service that can be used by any gui. That way X, gnome and KDE program would all do the same right thing.
Version after version, distro after distro, app after app, I still don't have a unified clipboard to simply copy and paste between apps.
You may find this funny, but I became a Linux coder last years after a long time on windows. To me this clipboad fiasco is both frustrating and symptomatic of what's wrong with windows.
I'm mostly coding using xemacs, but to ward off flamewars I also drop into vi from time to time.
First, what's wrong:
There is no unified clipboard. I can to Ctrl-k ctrl-y to cut and paste in Xemacs, but it doesn't work across different xemacs windows. I can use the Xemacs toolbar buttons labeled "cut", "copy" and "paste". These do work across xemacs instances, but do not interact with the ctrl-k ctrl-y buffers, and if I should close the source window before pasting, the copied text vanishes.
Then there is good-old x-windows swip & middle-click. This works in most everything, but is completely unrelated to the other two mechanisms. I use 3 different, incompatible, uninteroperable clipboards on a daily basis. Unix = simple orthogonal tools my arse.
A solution: Have a service/damon that stores ckips. It doesn't need to be part of any gui. It doesn't need to be linux-specific, it can be generic unix, and run on BSD too. It just needs a simple API (yes, like windows has) that supports multiple clips, perhaps taged by creating app and data type. Give it lots of config options like max size & number of clips to store, if it should persist them on HD when the system shuts down, query functions to find out what's there etc.
This, for me would be great. Apps that guard thier own clipboard jeolously could use thier own space on the clip server. Apps that want to look at other clips could do so. You could cut in KDE, restart and paste in Gnome.Heck, you could cut in emacs and paste in vi. It would be nerdvana. Why is this so hard?
This works now
on
Peek-a-Boo(ty)
·
· Score: 4, Insightful
the only way to connect... is word-of-mouth, which is horribly ineffective. Finding... will be extremely difficult unless you know the right people, and then it's very easy for the censor to ruin it. Trust the wrong person, and your whole network is exposed.
Millions of drug users use this model quite happily.
MOD blinkered rant up please
on
What is .NET?
·
· Score: 2
Flamebait? maybe. But the title "Blinkered rant" takes a lot of the sting out of that. Real flamebait tries to catch you unawares.
Moderators, please mod the parent up. Sure it's an extreme opinion. This too is good if it is acknoledged as such. Seldom have I seen a post that manages to be so funny, insightfull, full of bile and honest all at the same time.
How many of these people are MBA's vice-presidents of marketing or business analysts.
They don't mention anything about out-of-work programmers, sysadmins and webmasters. I'd think that a lower percentage of real techies are out of work.
So, you want feature X? Put it in that box - we'll look at it after we release.
I agree with using that approach when needed, I have even used it on occation,
BUT
You do realise that this is something that management should do all by themselves, and if the programmers are having to argue for that with the managers, then the managers are not doing that aspect of thier job.
A boss who leans over your shoulder every five minutes and gives misguided advice is a useless, but Leave me alone until I'm done is equally fatal. The majoroity of programming projects are characterised by deadline overruns.
There is a happy medium. "management is about balance" as someone else remarked in here.
Or, "don't go dark" as another project management guru remarked: If your project is in an unknown state, get it into a known state as soon as posible. Knowing that you are overtime is always better than not knowing.
My ideal boss in this case would:
- Orgainsise so that there is a real-world deliverable that will get used and bring feedback from end-users within six months.
- Help set milestones towards that occur regularly on the way to that goal.
- Make sure that activities besides coding take place, such as QA, code reviews, a modicum of design, documentation, etc.
- Ensure that the users are consulted so that what they get initially is vaugly usefull to them.
- Have a development meeting roughly once per week so that we can see if we are meeting our milestones, and if not, then revise our schedule, throw out features, or otherwise make sure that we are still in contact with reality.
Are you that dense?
Sometimes. IMHO this debate has become a silly excerse in hairsplitting and points of view.
When will you realize that Microshaft very rarely has a "NEW" idea?!?!?!?!?!
1) I did not claim they they had. "Borland did not do X" does not imply in any way that "X is Microsoft's idea".
2) Microsoft is a company. Companies do not have ideas. People do. Microsoft employs, truth be told, some of the best and brightest people that money can buy. Add it up. You and I may not like that but it is the truth. For instance, I have a lot of respect for Anders Hejlsberg, Archtect of Broland Turbo Pascal, Borland Delphi and Microsoft C#
You said they don't use an intermediate language of any kind, but as you should be able to see from the above text, they actually do. The difference is Borland kept this idea in the background.
There are important distinctations that you are not making. All the languages in.NET compile down to a common intermediate language, which is the final format of the code, and is interoperable between languages.
Delphi and BCB both finally compile down to x86 machine code. Sure the two compilers share code in the last pass of this process but so what - the common intermediate stage is not exposed, it is purely a time-saving device to save borland from writing a second compiler backend, no Delphi or BCB program ever runs in this intermediate language
I repeat, both Delphi and BCB generate x86 machine code, and.NET is fundamentally different from this in that it has a higher-level source-language independant IL which is run by the virtual machine. Borland has never had this.
Right. A proof of concept was when other languages were compiled to java bytecode. There may have been earlier ones too.
If I remember correctly is used in Borlands C++ Builder & Delphi to some extent. C++ Builder compilers down to the same thing as Delphi and if I remember correctly they use the same linker.... I beleive this is where they got this ideas behind CIL and combined it with the Java virtual Machine model.
Um. Nope. CIL stands for Common Intermediate Language.
Neither Delphi nor BCB are interpreted, JIT'd or run on virtual machine. Nor do they use an intermediate language of any kind.
The language that they have in common is in fact x86 assembly. The reason that the Delphi VCL can be used in BCB is that they share an object file format.
Maybe the idea of sharing a class library came over from Borland with Anders Heijsberg, but the method of achieving it is different.
What does it matter if Linus is using a source control system if no one else has access to it? I think that's really the whole point of using such a system, isn't it? So that multiple developers can check their code in manageably?
Nope. Version and revision tracking are also very important.
The article is about how a technology that geeks could do is now going mainstream. Thier product is an attempt to make a mass-market PC-video solution that a non-geek can use, with consumer bells and whisles like downloading TV guide listings from the web, software bundled with TVcard hardware, scheduled recording, etc. If they did thier work right, it should have a point-and-drool interface.
And the article does have a point. When a few geeks trade thier favorite show, it's no big loss. When everyone and thier Aunt Sally does, the media industry is in the acid bath.
There's plenty of change tracking capability. They call it the CHANGELOG.
A changlog (e.g. "Fixed stuff in file foo.c - me) is no substitute for the ability to compare any arbitrarily selected versions of the file and see every single difference in the source. If you've used source control systems like you say, you'll know this.
Please tell me that CVS server isn't on the same machine you just wrote the code on.
Usually on Sourceforge, but sometimes just on a different HD. The source is always on at least 2 different machines, even in the CVS repository is only on one of them.
CVS - I don't blame Linus one bit for choosing to stay far away from it
yes it can be crufty. Maybe Linus shouldn't use that particular source control system. But that's missing the point in a big way.
Microsoft SourceSafe is the worst
My experience is that the UI is slick, version five has engine problems, but V6 is generally OK, but the functionaity is limited, but if you are only doing straightforward stuff and like paying for closed-source software that only works on Windows, with an undocmented and highly obfuscated backend storage format, it will suffice. (Suprisingly, this describes very many IT shops.)
Your millage may vary.
Ok it's got to be said. In the words of whoever "source control is like brushing your teeth: you don't have to do it to all of them, just the ones that you want to keep."
IMHO (very humble in this case as far al kernel hacking goes, but I have been writing software for a decade, using various source control systems for half of that) the fact that Linus doesn't use any kind of source control/version tracking system is immature, stupid, counterproductive and just plain wrong on his part. It's shortsighted that a project of that size and importance depends on the buddy-system (ie Alan Cox has the source as well, etc..) for backups and revision tracking. When I write a few 100's of lines of utility code that i intend to keep, you bet it's in CVS!
No, CVS or other source control is not a magic bullet, you have to use it right. But if you don't have one, you don't even have a gun.
I'm not saying he should have a CVS tree with every joe haxor and their dog allowed to write to it. But no version history and change tracking capability at all!Gibber!
I can assure you its not the case that they meant to bundle this
So how is that relevant? If I drive my car into someone and kill them, but I was asleep at the wheel, does that mean that I am therefor innocent of any wrongdoing? Nope.
After reading the wired article, I think its pretty understandable how this slipped past the guys at Audiogalaxy.
I say judge them by their deeds not thier intensions - Audiogalaxy is in the business of distibuting software. How the crap can they not know what they are distributing? And if that is truly the case, it is thier problem.
She added that Borland plans to announce C++Builder for Windows next month, but declined to provide details.
That's got to be a journalist mangling the message. Borland has been selling C++ Builder for windows for several years already - why would they anounce it now?
This could really mean that either there is a new version of BCB for win32 to be anounced, or that BCB for linux is coming out.
Now it has been Borland's stated intent that the Kylix product line (or even the Kylix product) will support drag&drop C++ as well as Object-Pascal. I don't see why not - they share a compiler back end and a class library.
So either
- Borland has changed direction and not told anyone, and there will be no BCB for linux.
- We are getting BCB for Linux now.
- This is in fact only a commandline compiler, and is just a warmup for a later release of BCB for Linux.
Nothing beats having what I want, when I want, for how much I want it!
And do you want to take an hour driving to work in the morning at 5 mph? Thought not. Learn to understand that you are not alone on this planet, that you gain from cooperation, that that you are not stuck in the trafic jam, you are part of the trafic jam. Then we might begin to solve these problems, and you might get something better than what you have now.
Slashdot Mirror
Right, and if clipping was recognised as nothing to do with a GUI, just a bit of data storage, then TTY apps could use it.
It sounds to me as though you are starting multiple independent instances of Xemacs to edit multiple files - I think that you will find that running a single Emacs process and opening multiple "frames" to edit multiple buffers is the correct way to go.
I've seen those frames and don't like them much. If you don't have a windowing system, then I suppose that they would be a necessary substitute.
The basic problem is that different instances of the same program, let alone different ones, don't play nice together.
That. The mechanics of storing clips has got nothing to do with a gui of any kind, X or otherwise. It should be independant of X, ie usable from commandline vi and emacs.
And that's where the problems start. Thee is nothing about storing smallish pieces of data temporarily that needs to be part of any gui or windowing toolkit - it's not a visual thing in any way. It should be a system service that can be used by any gui. That way X, gnome and KDE program would all do the same right thing.
You may find this funny, but I became a Linux coder last years after a long time on windows. To me this clipboad fiasco is both frustrating and symptomatic of what's wrong with windows.
I'm mostly coding using xemacs, but to ward off flamewars I also drop into vi from time to time.
First, what's wrong:
There is no unified clipboard. I can to Ctrl-k ctrl-y to cut and paste in Xemacs, but it doesn't work across different xemacs windows. I can use the Xemacs toolbar buttons labeled "cut", "copy" and "paste". These do work across xemacs instances, but do not interact with the ctrl-k ctrl-y buffers, and if I should close the source window before pasting, the copied text vanishes. Then there is good-old x-windows swip & middle-click. This works in most everything, but is completely unrelated to the other two mechanisms. I use 3 different, incompatible, uninteroperable clipboards on a daily basis. Unix = simple orthogonal tools my arse.
A solution:
Have a service/damon that stores ckips. It doesn't need to be part of any gui. It doesn't need to be linux-specific, it can be generic unix, and run on BSD too. It just needs a simple API (yes, like windows has) that supports multiple clips, perhaps taged by creating app and data type. Give it lots of config options like max size & number of clips to store, if it should persist them on HD when the system shuts down, query functions to find out what's there etc.
This, for me would be great. Apps that guard thier own clipboard jeolously could use thier own space on the clip server. Apps that want to look at other clips could do so. You could cut in KDE, restart and paste in Gnome.Heck, you could cut in emacs and paste in vi. It would be nerdvana. Why is this so hard?
Millions of drug users use this model quite happily.
Moderators, please mod the parent up. Sure it's an extreme opinion. This too is good if it is acknoledged as such. Seldom have I seen a post that manages to be so funny, insightfull, full of bile and honest all at the same time.
I wouldn't go that far. The
The slash engine was designed? I thought it just sort of grew. Like a fungus.
Like some other software out there. I qoute:
"Software "Design" is like a religion - too much of it makes you inflexibly and unpopular."
- Linus Torvalds
Someone's got to say it:
How many of these people are MBA's vice-presidents of marketing or business analysts.
They don't mention anything about out-of-work programmers, sysadmins and webmasters. I'd think that a lower percentage of real techies are out of work.
Replies welcome any out-of-work C coders. Anyone?
I agree with using that approach when needed, I have even used it on occation,
BUT
You do realise that this is something that management should do all by themselves, and if the programmers are having to argue for that with the managers, then the managers are not doing that aspect of thier job.
There is a happy medium. "management is about balance" as someone else remarked in here.
Or, "don't go dark" as another project management guru remarked: If your project is in an unknown state, get it into a known state as soon as posible. Knowing that you are overtime is always better than not knowing.
My ideal boss in this case would:
- Orgainsise so that there is a real-world deliverable that will get used and bring feedback from end-users within six months.
- Help set milestones towards that occur regularly on the way to that goal.
- Make sure that activities besides coding take place, such as QA, code reviews, a modicum of design, documentation, etc.
- Ensure that the users are consulted so that what they get initially is vaugly usefull to them.
- Have a development meeting roughly once per week so that we can see if we are meeting our milestones, and if not, then revise our schedule, throw out features, or otherwise make sure that we are still in contact with reality.
Sometimes. IMHO this debate has become a silly excerse in hairsplitting and points of view.
When will you realize that Microshaft very rarely has a "NEW" idea?!?!?!?!?!
1) I did not claim they they had. "Borland did not do X" does not imply in any way that "X is Microsoft's idea".
2) Microsoft is a company. Companies do not have ideas. People do. Microsoft employs, truth be told, some of the best and brightest people that money can buy. Add it up. You and I may not like that but it is the truth. For instance, I have a lot of respect for Anders Hejlsberg, Archtect of Broland Turbo Pascal, Borland Delphi and Microsoft C#
There are important distinctations that you are not making. All the languages in .NET compile down to a common intermediate language, which is the final format of the code, and is interoperable between languages.
Delphi and BCB both finally compile down to x86 machine code. Sure the two compilers share code in the last pass of this process but so what - the common intermediate stage is not exposed, it is purely a time-saving device to save borland from writing a second compiler backend, no Delphi or BCB program ever runs in this intermediate language
I repeat, both Delphi and BCB generate x86 machine code, and .NET is fundamentally different from this in that it has a higher-level source-language independant IL which is run by the virtual machine. Borland has never had this.
Right. A proof of concept was when other languages were compiled to java bytecode. There may have been earlier ones too.
If I remember correctly is used in Borlands C++ Builder & Delphi to some extent. C++ Builder compilers down to the same thing as Delphi and if I remember correctly they use the same linker. ... I beleive this is where they got this ideas behind CIL and combined it with the Java virtual Machine model.
Um. Nope. CIL stands for Common Intermediate Language.
Neither Delphi nor BCB are interpreted, JIT'd or run on virtual machine. Nor do they use an intermediate language of any kind.
The language that they have in common is in fact x86 assembly. The reason that the Delphi VCL can be used in BCB is that they share an object file format.
Maybe the idea of sharing a class library came over from Borland with Anders Heijsberg, but the method of achieving it is different.
We have a live system running on postgres 7.1 Now what about upgrading?
When should we do it? Should we wait a couple of weeks for any possible patch/point release, or is this one granite-rock-solid?
Why should we do it? The current DB is working just fine. Are there any compelling new features?
How do we do it? Are the database file binary compatible? Do we need to export and reimport data? Are the RPMs of postgres 7.2 out?
Nope. Version and revision tracking are also very important.
> Putting the Linux kernel on CVS (or Arch or whatever) would just allow people to commit stupid changes.
Not true. Putting the Linux kernel on publically writable CVS would allow that, but no one is seriously suggesting that.
RTFA
The article is about how a technology that geeks could do is now going mainstream. Thier product is an attempt to make a mass-market PC-video solution that a non-geek can use, with consumer bells and whisles like downloading TV guide listings from the web, software bundled with TVcard hardware, scheduled recording, etc. If they did thier work right, it should have a point-and-drool interface.
And the article does have a point. When a few geeks trade thier favorite show, it's no big loss. When everyone and thier Aunt Sally does, the media industry is in the acid bath.
A changlog (e.g. "Fixed stuff in file foo.c - me) is no substitute for the ability to compare any arbitrarily selected versions of the file and see every single difference in the source. If you've used source control systems like you say, you'll know this.
Please tell me that CVS server isn't on the same machine you just wrote the code on.
Usually on Sourceforge, but sometimes just on a different HD. The source is always on at least 2 different machines, even in the CVS repository is only on one of them.
CVS - I don't blame Linus one bit for choosing to stay far away from it yes it can be crufty. Maybe Linus shouldn't use that particular source control system. But that's missing the point in a big way.
Microsoft SourceSafe is the worst My experience is that the UI is slick, version five has engine problems, but V6 is generally OK, but the functionaity is limited, but if you are only doing straightforward stuff and like paying for closed-source software that only works on Windows, with an undocmented and highly obfuscated backend storage format, it will suffice. (Suprisingly, this describes very many IT shops.) Your millage may vary.
IMHO (very humble in this case as far al kernel hacking goes, but I have been writing software for a decade, using various source control systems for half of that) the fact that Linus doesn't use any kind of source control/version tracking system is immature, stupid, counterproductive and just plain wrong on his part. It's shortsighted that a project of that size and importance depends on the buddy-system (ie Alan Cox has the source as well, etc..) for backups and revision tracking. When I write a few 100's of lines of utility code that i intend to keep, you bet it's in CVS!
No, CVS or other source control is not a magic bullet, you have to use it right. But if you don't have one, you don't even have a gun.
I'm not saying he should have a CVS tree with every joe haxor and their dog allowed to write to it. But no version history and change tracking capability at all! Gibber!
But. A policy of including stuff in your product when you Don't know what it does is just wrong and exposes you to liability.
So how is that relevant? If I drive my car into someone and kill them, but I was asleep at the wheel, does that mean that I am therefor innocent of any wrongdoing? Nope.
After reading the wired article, I think its pretty understandable how this slipped past the guys at Audiogalaxy.
I say judge them by their deeds not thier intensions - Audiogalaxy is in the business of distibuting software. How the crap can they not know what they are distributing? And if that is truly the case, it is thier problem.
Now "development solution" doesn't sound like just a commndline compiler to me.
I expect that option 2 (BCB for linux now) is most likely, and option 1 (No BCB for linux ever) is right out.
There will likely also be a new version of bcb for win32. That's what happened on the Object-Pascal front: Kylix 1 rapidly followed by Delphi6.
That's got to be a journalist mangling the message. Borland has been selling C++ Builder for windows for several years already - why would they anounce it now?
This could really mean that either there is a new version of BCB for win32 to be anounced, or that BCB for linux is coming out.
Now it has been Borland's stated intent that the Kylix product line (or even the Kylix product) will support drag&drop C++ as well as Object-Pascal. I don't see why not - they share a compiler back end and a class library.
So either
- Borland has changed direction and not told anyone, and there will be no BCB for linux.
- We are getting BCB for Linux now.
- This is in fact only a commandline compiler, and is just a warmup for a later release of BCB for Linux.
Place your bets.
And do you want to take an hour driving to work in the morning at 5 mph? Thought not. Learn to understand that you are not alone on this planet, that you gain from cooperation, that that you are not stuck in the trafic jam, you are part of the trafic jam. Then we might begin to solve these problems, and you might get something better than what you have now.