Eclipse is fast, GTK native, full open-source with a very well done plugin architecture.
Point by point:
Fast: well, it's all relative, I suppose. For example, according to Einstein, if you were to speed around the world in a fraction of a second and end up back at your computer seat, maybe it would actually accomplish something (e.g., actually display the key I typed on my keyboard) in a under a second. But let's call a spade a spade here: Eclipse is one of the slowest (useful) programs in the world. Oh, and ever try the Visual Editor? Makes the rest of Eclipse look speedy.
GTK native: well, that all depends on your definition of "native". You could say Mozilla uses native GTK widgets, but that doesn't stop its horizontal scroll bar from behaving funny. Just like Eclipse's tabs and docking features behave in approximately the exact opposite manner of the rest of my GNOME applications.
Plugin architecture: yes, yes, congrats, it's great. Until you try to change a preference and all those plugins jump out at you simultaneously.
Don't get me wrong. I think Eclipse is hands-down the best Java IDE in the world. But it's the only half-decent Java IDE, period. It's also the least consistent with the rest of my desktop, and it's way too bloated.
Yes, Evince is absolutely fantastic. The fullscreen, find, and thumbnail features set it apart from any other PDF viewer. And its load time is absolutely incomparable to any other PDF viewer. And its interface is simpler. And... and... and....
After an experience with Evince, I can't imagine going back to Acrobat. I can't even imagine going back to xpdf.
I love the implicit abuse of the scientific method in this article:
Results: "People with high working memories are more affected by pressure than those with low working memories" (by the way, a sample of 97 people really isn't that conclusive).
* IIS? Gone with Apache.
* File/Print? Gone with Samba.
* Email? Not so fast. We like the groupware functionality of Exchange.
Utter bullshit. Sunbird isn't appropriate as a small business integrated solution. Why am I so sure? Because it's not integrated with anything!
Small business already have no reason to use Windows; Novell (Ximian, SuSE) and Red Hat have had comprehensive solutions for years. Evolution already works way better than Outlook.
I'm not going to go as far as to say that Sunbird is a waste of effort -- it hasn't yet proven itself to be otherwise, but that doesn't mean it won't -- I'm just saying there is software written specifically to fit the niche you describe, and it is not Sunbird. Open source is all about "right tool for the job", after all.
They obviously won't be using valid (and correctly-served) XHTML 1.1, since IE refuses to even render application/xhtml+xml documents as XHTML.
And judging by previous comments, they're not even bothering with XHTML 1.0, either. And writing invalid XHTML 1.0 is much worse than writing invalid HTML, since XHTML is XML and should thus die horribly if there's a single error.
Out of curiosity, why does Microsoft find it so hard to write valid XHTML when everyone else finds it so easy? And in general... why the FUCK don't full-time web developers write valid XHTML? AARGH!
From what I can understand in the article, XML is just used to represent an abstract syntax tree with metadata (e.g., "<doc>" tags). The author even admits that we could have done the same thing with LISP expressions....
If the author's suggested systems were to come about (and they certainly sound nice), why on Earth would they use XML?
The compiler, debugger, and editor (plus profiler and other helper apps) would all have to be written for the new data storage format. Without one, the entire system will not work. It stands to reason that this work will be done by one group or company. So they can define whatever data format they want. Why XML? "XSLT transforms" is a lousy answer: transforms would be trivial to implement in a scriptable manner no matter what storage format is used to represent the AST.
And while providing debugging methods in a library is a fantastic idea, it really has nothing to do with XML at all. It would just be a hack on the ELF (or whatever) binary format which the compiler would be able to deal with.
The fact is, nothing in this article really gives a reason to use XML. Maybe it would be a nice interface between the editor and the compiler front-end, but beyond that there's really no reason at all....
Actually, I'll back up another step: if you've got an editor which stores class diagrams in its comments, that's one mother of an editor. It stands to reason it's an IDE which invokes the compiler for you automatically. Writing an object or two in the editor's code that translate its internal representation into C (or whatever) would be trivial compared to the rest of the IDE. So why bother even changing the compiler?
This article basically boils down to, "All IDEs suck. There ought to be a good one." Even using the GNU tools as they are today, a killer IDE could be written to implement all the features suggested. And XML wouldn't be needed at all.
we, users, get functionality, which idea is best is proven by real world, not by marketing crap
No, that's not really true. Storage could have been great, but had only one person working on it. For free. And thus not all that much. As for Beagle: look through its AUTHORS file and you'll see an awful lot of @ximian.org's and @novell.com's. These guys are getting paid to hack.
In this instance, the end result seems to be good. In fact, there actually is a pretty nice "look how cool open-source is" story. Before Beagle, there was Dashboard (an always-open window which is supposed to give you links based on what you're doing -- e.g., who you're talking to in Gaim). Dashboard simply never took off. Then somebody came up with the idea of turning it into a search tool.
As an Epiphany fan, though, I simply can't agree with that comment about open-source not being affected by "marketing crap". Ubuntu's choice of Firefox as default browser was because of its brand name recognition. (Yes, I can back that up.)
Anyway, my point is: "open source" isn't really a panacea for great software. Free Software isn't free from the real world. The open source world isn't perfect; it's just better than the proprietary one.
Epiphany has better extensions support
on
Gnome 2.10 Sneak Peek
·
· Score: 4, Informative
Unmentioned on that page: Epiphany extensions can now be loaded/unloaded on-the-fly. The epiphany-extensions package comes with an extension which lets you do this. And the adblock extension is coming, dammit!
And there's also "pyphany" in CVS. It lets you make extensions using Python. Included in the CVS module: a Python Console extension, which is probably the best way to prototype extensions (you can, say, connect a signal to change the zoom, with just a couple of lines of code).
Anyone knows if, say, Evince is going the way kpdf is?
I don't know how kpdf is, but I know that Evince is already awesome. It thumbnails pages in the left pane; has a Fullscreen mode (copied from Epiphany, I think -- Marco is working on Evince); supports hyperlinks and indexes; has a Firefox-esque Find bar which actually works; and it loads really, really, REALLY fast. (But it's only available in CVS for now, I think.)
You don't have a field where you can type/paste a file path
Ctrl-L.
half the time when you hit "Save" you actually have to click another control in order to be able to save it anywhere other than the current folder or a few fairly retarded locations like/home or /...
Yeah, and the other half of the time it's marvellously fast. You're probably actually spending *less* time with the new dialogs (using Ctrl-L) than you were with the old ones -- even if you have to type in file paths -- because half of the time, you don't.
And you can set yourself bookmarks, if you'd like. Just drag folders onto the left pane and they'll show up in the drop-down next time you want to save a file.
Teaching people to understand how file structures work in an intuitive manner is one of the functions of a file open/save dialog
No. Hiding the details of file structures from users is one of the functions of a file open/save dialog. Most users don't understand what a file hierarchy is... and don't want to.
And, also - this bugs me... why on earth does Gnome have a Windows-style registry?
I challenge you to name three significant architectural similarities between GConf and the WIndows registry. They are entirely different.
Actually, GNOME Storage is a pretty dead project. What people probably want to see screenshots for these days is Beagle. Beagle gathers metadata and indexes content instead of replacing the filesystem. And it Just Works. Has done so for months.
the common non-tech-savvy person often will choose a version of Windows over a distro of Linux simply because XP just looks nicer and easier to use.
GTK+ is fully themeable, but the default GNOME theme has many restrictions -- basically, it has to look good for everybody. And as others have mentioned here, the GNOME theme is actually more "useful" -- less space taken up by big blue curvy things.
For reference: the default theme actually did change in GNOME 2.8 (and my, was the older theme ever outdated). For an overview of the design issues, check out this mailing list thread.
Any other libraries you need, you just take your own copy with you in your application bundle and call it from there. Sure, you might end up with duplicate copies of libraries, but who gives a crap? Disk space is cheap as hell.
When you run two programs at once (e.g., Thunderbird and Firefox), that disk space turns into memory space. What's worse: when you upgrade one instance of the library and not the other, security fixes aren't carried over and inconsistencies can arise between programs.
For example, recent versions of GTK+ have dramatically improved minor details in the new file selector (I'm not talking of the shift from old to new, I'm talking of keyboard shortcuts, bookmark handling, etc). If not using shared libraries, the file selector would behave differently in different applications.
The "Program Files" directory paradigm makes sense in a commercial environment, since it's difficult to make your program co-operate with others'. But in the Free Software world, the opposite is true: if there's a bug in some library, fix the library instead of creating a workaround in your program. Code stays cleaner and all programs dependent upon that library benefit. It's a net gain.
So when dealing with Free Software, the package manager works perfectly (except when people assume they know better than the package manager). It's a far better solution than statically-linked programs.
There is a time for learning for learnings sake - retirement.
That's one way of looking at it, sure. But I think I'll learn for learning's sake my entire life, thank you very much. That way I wouldn't feel my life was a waste of time if I died at 64.
It's a little like locking a house: if you try to make the locking system as unobtrusive and inconspicuous as possible, somebody will test to see if it works, but if the system appears extremely difficult to break down as soon as you look at it, anyone with evil intentions will probably admit that it's not even worth trying to break in there.
This is exactly the opposite of Paul Graham's attitude expressed in the Good Bad Attitude mentioned on Slashdot recently:
Show any hacker a lock and his first thought is how to pick it.
I make it no secret that I sleep with my apartment door unlocked. Nobody has ever "tested" my door, despite the facts that I'm likely asleep and there's likely expensive stuff inside. However, were I to install a large array of deadbolts positioned very visibly on the outside of my door, I feel I'd likely have somebody stare at them for a while, wondering just how to get at what's on the other side.
I agree with Paul Graham to a much greater extent than Hannu H. Kari.
A normal building fire would not have threatened the structural integrity of the WTC because there was nothing in the WTC that was hot enough to melt the beams, until the plane, full of fuel, arrived.
Not only that. The beams had some stuff sprayed on them to shield them from heat. The impact of the planes removed the heat shielding, leaving the beams exposed to the heat from the fuel.
slackware, debian, gentoo, fedora, suse, mandrake, and maybe a few others that offer something good, that's all we really need.
From my understanding, Ubunto is exactly like Debian except it's (1) end-user centered and (2) released in a timely manner.
These two things are exactly what I want! Seems to me like the perfect blend of GNOME and Debian ideals. Thus, if I get up the courage to try it out and Ubunto delivers, I will likely be a convert.
4) Debian makes me feel guilty if I want to use any reasonably up-to-date software.:) I have to use SUPER-UNSTABLE-OMFG-ITS-GONNA-BREAK distro. So many posts around insist "don't use Sid, it'll break! OMFG!!11 use Sarge!".
Yes, "stable" in Debian is exactly what you'd expect. It's solid as a rock, but it's as exciting as one, too.
In my experience, Gentoo's "stable" is about on par with Debian's "unstable" branch in terms of stability (i.e., upgrades breaking things, etc). That's not to say that Gentoo is unstable -- simply that Debian "unstable" isn't anything to be afraid of. In the past 6 months I've had one bug appear from a dist-upgrade (the "alt-tab" bug, I think Gentoo had it too).
Just thought I'd add a quick plug for Epiphany Extensions. We worked on a couple (Page Info and Select Stylesheet) right before the deadline, so now we've got a somewhat reasonable bunch.
Epiphany is still a browser centred around simplicity. But the extensions can give you those features you wish you had from other browsers.
The full list: SSL certificate viewer, dashboard connection, HTML/Javascript error viewer and link checker, mouse gestures, page info dialog, stylesheet selection, "smart bookmarks" (right-click on selection -> search the web), tab grouping (open new tabs directly next to the current one), tabs menu entries.
However, it's not until GNOME 2.10 that there'll be a UI to select extensions.
Whoa. The article says MD5 encryption was cracked. Is this true? I had only heard a rumour that a single collision had been found. And as for SHA-0... that's hardly making up the backbone of the Internet, is it?
Wait a minute... is MD5 even encryption at all? I thought it was a "message digest"...
Aside from training, the games also improve young people's perceptions of the military
How can one person's perception of the military be "better" than another's? Presumably if it's closer to reality, not necessarily more favourable. So I'm guessing the games do the exact opposite.
Slashdot Mirror
Eclipse is fast, GTK native, full open-source with a very well done plugin architecture .
Point by point:
Don't get me wrong. I think Eclipse is hands-down the best Java IDE in the world. But it's the only half-decent Java IDE, period. It's also the least consistent with the rest of my desktop, and it's way too bloated.
Yes, Evince is absolutely fantastic. The fullscreen, find, and thumbnail features set it apart from any other PDF viewer. And its load time is absolutely incomparable to any other PDF viewer. And its interface is simpler. And... and... and.... After an experience with Evince, I can't imagine going back to Acrobat. I can't even imagine going back to xpdf.
I love the implicit abuse of the scientific method in this article:
Results: "People with high working memories are more affected by pressure than those with low working memories" (by the way, a sample of 97 people really isn't that conclusive).
Conclusion: "Smart people choke under pressure".
* IIS? Gone with Apache.
* File/Print? Gone with Samba.
* Email? Not so fast. We like the groupware functionality of Exchange.
Utter bullshit. Sunbird isn't appropriate as a small business integrated solution. Why am I so sure? Because it's not integrated with anything!
Small business already have no reason to use Windows; Novell (Ximian, SuSE) and Red Hat have had comprehensive solutions for years. Evolution already works way better than Outlook.
I'm not going to go as far as to say that Sunbird is a waste of effort -- it hasn't yet proven itself to be otherwise, but that doesn't mean it won't -- I'm just saying there is software written specifically to fit the niche you describe, and it is not Sunbird. Open source is all about "right tool for the job", after all.
They obviously won't be using valid (and correctly-served) XHTML 1.1, since IE refuses to even render application/xhtml+xml documents as XHTML.
And judging by previous comments, they're not even bothering with XHTML 1.0, either. And writing invalid XHTML 1.0 is much worse than writing invalid HTML, since XHTML is XML and should thus die horribly if there's a single error.
Out of curiosity, why does Microsoft find it so hard to write valid XHTML when everyone else finds it so easy? And in general... why the FUCK don't full-time web developers write valid XHTML? AARGH!
From what I can understand in the article, XML is just used to represent an abstract syntax tree with metadata (e.g., "<doc>" tags). The author even admits that we could have done the same thing with LISP expressions....
If the author's suggested systems were to come about (and they certainly sound nice), why on Earth would they use XML?
The compiler, debugger, and editor (plus profiler and other helper apps) would all have to be written for the new data storage format. Without one, the entire system will not work. It stands to reason that this work will be done by one group or company. So they can define whatever data format they want. Why XML? "XSLT transforms" is a lousy answer: transforms would be trivial to implement in a scriptable manner no matter what storage format is used to represent the AST.
And while providing debugging methods in a library is a fantastic idea, it really has nothing to do with XML at all. It would just be a hack on the ELF (or whatever) binary format which the compiler would be able to deal with.
The fact is, nothing in this article really gives a reason to use XML. Maybe it would be a nice interface between the editor and the compiler front-end, but beyond that there's really no reason at all....
Actually, I'll back up another step: if you've got an editor which stores class diagrams in its comments, that's one mother of an editor. It stands to reason it's an IDE which invokes the compiler for you automatically. Writing an object or two in the editor's code that translate its internal representation into C (or whatever) would be trivial compared to the rest of the IDE. So why bother even changing the compiler?
This article basically boils down to, "All IDEs suck. There ought to be a good one." Even using the GNU tools as they are today, a killer IDE could be written to implement all the features suggested. And XML wouldn't be needed at all.
we, users, get functionality, which idea is best is proven by real world, not by marketing crap
No, that's not really true. Storage could have been great, but had only one person working on it. For free. And thus not all that much. As for Beagle: look through its AUTHORS file and you'll see an awful lot of @ximian.org's and @novell.com's. These guys are getting paid to hack.
In this instance, the end result seems to be good. In fact, there actually is a pretty nice "look how cool open-source is" story. Before Beagle, there was Dashboard (an always-open window which is supposed to give you links based on what you're doing -- e.g., who you're talking to in Gaim). Dashboard simply never took off. Then somebody came up with the idea of turning it into a search tool.
As an Epiphany fan, though, I simply can't agree with that comment about open-source not being affected by "marketing crap". Ubuntu's choice of Firefox as default browser was because of its brand name recognition. (Yes, I can back that up.)
Anyway, my point is: "open source" isn't really a panacea for great software. Free Software isn't free from the real world. The open source world isn't perfect; it's just better than the proprietary one.
Unmentioned on that page: Epiphany extensions can now be loaded/unloaded on-the-fly. The epiphany-extensions package comes with an extension which lets you do this. And the adblock extension is coming, dammit!
And there's also "pyphany" in CVS. It lets you make extensions using Python. Included in the CVS module: a Python Console extension, which is probably the best way to prototype extensions (you can, say, connect a signal to change the zoom, with just a couple of lines of code).
Anyone knows if, say, Evince is going the way kpdf is?
I don't know how kpdf is, but I know that Evince is already awesome. It thumbnails pages in the left pane; has a Fullscreen mode (copied from Epiphany, I think -- Marco is working on Evince); supports hyperlinks and indexes; has a Firefox-esque Find bar which actually works; and it loads really, really, REALLY fast. (But it's only available in CVS for now, I think.)
Not if the Enlightenment guys manage to do a release.
The words "Hell" and "freezing over" come to mind. Isn't GNOME's 6-month release cycle great!
You don't have a field where you can type/paste a file path
Ctrl-L.
half the time when you hit "Save" you actually have to click another control in order to be able to save it anywhere other than the current folder or a few fairly retarded locations like /home or / ...
Yeah, and the other half of the time it's marvellously fast. You're probably actually spending *less* time with the new dialogs (using Ctrl-L) than you were with the old ones -- even if you have to type in file paths -- because half of the time, you don't.
And you can set yourself bookmarks, if you'd like. Just drag folders onto the left pane and they'll show up in the drop-down next time you want to save a file.
Teaching people to understand how file structures work in an intuitive manner is one of the functions of a file open/save dialog
No. Hiding the details of file structures from users is one of the functions of a file open/save dialog. Most users don't understand what a file hierarchy is... and don't want to.
And, also - this bugs me... why on earth does Gnome have a Windows-style registry?
I challenge you to name three significant architectural similarities between GConf and the WIndows registry. They are entirely different.
Actually, GNOME Storage is a pretty dead project. What people probably want to see screenshots for these days is Beagle. Beagle gathers metadata and indexes content instead of replacing the filesystem. And it Just Works. Has done so for months.
the common non-tech-savvy person often will choose a version of Windows over a distro of Linux simply because XP just looks nicer and easier to use.
GTK+ is fully themeable, but the default GNOME theme has many restrictions -- basically, it has to look good for everybody. And as others have mentioned here, the GNOME theme is actually more "useful" -- less space taken up by big blue curvy things.
For reference: the default theme actually did change in GNOME 2.8 (and my, was the older theme ever outdated). For an overview of the design issues, check out this mailing list thread.
Any other libraries you need, you just take your own copy with you in your application bundle and call it from there. Sure, you might end up with duplicate copies of libraries, but who gives a crap? Disk space is cheap as hell.
When you run two programs at once (e.g., Thunderbird and Firefox), that disk space turns into memory space. What's worse: when you upgrade one instance of the library and not the other, security fixes aren't carried over and inconsistencies can arise between programs.
For example, recent versions of GTK+ have dramatically improved minor details in the new file selector (I'm not talking of the shift from old to new, I'm talking of keyboard shortcuts, bookmark handling, etc). If not using shared libraries, the file selector would behave differently in different applications.
The "Program Files" directory paradigm makes sense in a commercial environment, since it's difficult to make your program co-operate with others'. But in the Free Software world, the opposite is true: if there's a bug in some library, fix the library instead of creating a workaround in your program. Code stays cleaner and all programs dependent upon that library benefit. It's a net gain.
So when dealing with Free Software, the package manager works perfectly (except when people assume they know better than the package manager). It's a far better solution than statically-linked programs.
There is a time for learning for learnings sake - retirement.
That's one way of looking at it, sure. But I think I'll learn for learning's sake my entire life, thank you very much. That way I wouldn't feel my life was a waste of time if I died at 64.
The only time this would be a problem is when a whole new file is being written.
That's the second-most-common operation on file servers (the actual reading of files being the most common).
From the article:
It's a little like locking a house: if you try to make the locking system as unobtrusive and inconspicuous as possible, somebody will test to see if it works, but if the system appears extremely difficult to break down as soon as you look at it, anyone with evil intentions will probably admit that it's not even worth trying to break in there.
This is exactly the opposite of Paul Graham's attitude expressed in the Good Bad Attitude mentioned on Slashdot recently:
Show any hacker a lock and his first thought is how to pick it.
I make it no secret that I sleep with my apartment door unlocked. Nobody has ever "tested" my door, despite the facts that I'm likely asleep and there's likely expensive stuff inside. However, were I to install a large array of deadbolts positioned very visibly on the outside of my door, I feel I'd likely have somebody stare at them for a while, wondering just how to get at what's on the other side.
I agree with Paul Graham to a much greater extent than Hannu H. Kari.
A normal building fire would not have threatened the structural integrity of the WTC because there was nothing in the WTC that was hot enough to melt the beams, until the plane, full of fuel, arrived.
Not only that. The beams had some stuff sprayed on them to shield them from heat. The impact of the planes removed the heat shielding, leaving the beams exposed to the heat from the fuel.
Am I the only one who started hearing alarms going off in my head when I read this sentence:
"We probably have the most secure system in the nation," said Lamone...
Translation: "We know nothing about security."
And lo and behold, they're using Microsoft Access. I rest my case.
slackware, debian, gentoo, fedora, suse, mandrake, and maybe a few others that offer something good, that's all we really need.
From my understanding, Ubunto is exactly like Debian except it's (1) end-user centered and (2) released in a timely manner.
These two things are exactly what I want! Seems to me like the perfect blend of GNOME and Debian ideals. Thus, if I get up the courage to try it out and Ubunto delivers, I will likely be a convert.
4) Debian makes me feel guilty if I want to use any reasonably up-to-date software. :) I have to use SUPER-UNSTABLE-OMFG-ITS-GONNA-BREAK distro. So many posts around insist "don't use Sid, it'll break! OMFG!!11 use Sarge!".
Yes, "stable" in Debian is exactly what you'd expect. It's solid as a rock, but it's as exciting as one, too.
In my experience, Gentoo's "stable" is about on par with Debian's "unstable" branch in terms of stability (i.e., upgrades breaking things, etc). That's not to say that Gentoo is unstable -- simply that Debian "unstable" isn't anything to be afraid of. In the past 6 months I've had one bug appear from a dist-upgrade (the "alt-tab" bug, I think Gentoo had it too).
We're very keen on an adblock extension. Personally, I wouldn't write it unless I could think up a MUCH less sucky user interface.
Anyway, expect adblock in an upcoming epiphany-extensions release... probably before GNOME 2.10.
Just thought I'd add a quick plug for Epiphany Extensions. We worked on a couple (Page Info and Select Stylesheet) right before the deadline, so now we've got a somewhat reasonable bunch.
Epiphany is still a browser centred around simplicity. But the extensions can give you those features you wish you had from other browsers.
The full list: SSL certificate viewer, dashboard connection, HTML/Javascript error viewer and link checker, mouse gestures, page info dialog, stylesheet selection, "smart bookmarks" (right-click on selection -> search the web), tab grouping (open new tabs directly next to the current one), tabs menu entries.
However, it's not until GNOME 2.10 that there'll be a UI to select extensions.
Whoa. The article says MD5 encryption was cracked. Is this true? I had only heard a rumour that a single collision had been found. And as for SHA-0... that's hardly making up the backbone of the Internet, is it?
Wait a minute... is MD5 even encryption at all? I thought it was a "message digest"...
Aside from training, the games also improve young people's perceptions of the military
How can one person's perception of the military be "better" than another's? Presumably if it's closer to reality, not necessarily more favourable. So I'm guessing the games do the exact opposite.