There is already such a solution. It's called RFC 2369, and it dictates headers to be inserted into every message. It's existed for years; the problem is webmail and most mail client developers are too stupid or too selfish to implement support for these headers.
Please ignore my hostile attitude there as it is mostly directed at the anonymous post to which I was responding. In short, I am used to Pidgin's interface and don't like any other IM client's interface because of being so used to Pidgin.
As I have noted elsewhere, I am not a member of the Pidgin development team. I am a regular in the project's IRC channels, and I have contributed in the past, being responsible for the (long overdue) death of a large portion of the serv_* API namespace back in the earlier 2.0.0 development days among other contributions. Yes, I am hostile and standoffish toward Adium, and here's why:
I dislike its interface, as I have grown very accustomed to Pidgin's interface after having used Gaim/Pidgin since the days of 0.43.
I dislike the fact that when I try to combine multiple buddies into a "metacontact" in Adium, the buddy I have dragged and dropped becomes the priority buddy in the meta. This is just plain stupid. This is were Pidgin is superior, in that I can choose the buddy's priority in the contact when I create the contact instead of having to go to some stupid info dialog that should have NOTHING to do with the organization of a contact.
I dislike the fact that the accounts and preferences dialogs are one and the same in Adium.
I dislike the conversion of "/me does action" to "*does action*".
I dislike the tabbed conversation windows having their tabs along the bottom and not being able to change said placement.
And finally, I disagree on a fundamental level with the iTunes integration added to Adium. This is only an incentive for more abuse of status messages to display stupid crap like the currently playing song.
This is the first time I have ever publicly aired my complaints about Adium, and had hoped it would never happen. I will note that there are a few other reasons that I will never discuss with anyone, as a discussion of these issues will only cause more problems than it could ever solve. But, this being slashdot, the land of stupidity and flame wars, I was dragged in by a troll hook, line, and sinker and convinced to air some of my grievances. My only hope is that people are intelligent enough not to take my opinion as being that of any member of the Gaim/Pidgin development team that I am NOT officially affiliated with.
I'm not fully briefed on the decision process for monotone. I believe several of the developers felt monotone to be the most intuitive for handling conflicts and merges and such, but I could be mistaken. I know several of the developers, once trying monotone, darcs and one other whose name I can't recall at the moment (perhaps git or mercurial?), preferred monotone for their own use.
The names actually make sense. A pidgin is basically a mish-mash of languages, which is what we see on IM with all this "u", "o rly?", "omg" and similar crappy "abbreviations" floating around. It also sounds like "pigeon", which is where the new logo will be coming in eventually. Finch was named so to keep in with the bird theme, and I suspect to imply a "little brother" idea since it is young in comparison to Pidgin itself. Libpurple comes from the abbreviation long used for protocol plugins in the client, "prpl", which is supposed to be pronounced "purple." Since libpurple's major function is to provide the IM service handling backend for messaging applications, it made sense that the protocol plugin's abbreviation should have some part in its new name.
This is not saying that the developers don't want people hacking on Pidgin. There are currently plans in place to implement a Subversion gateway so that casual hackers can pull the Pidgin source and create their patches and whatnot. The reason behind the switch to monotone is that a distributed version control system fits more in line with the core developers' workflow, working on things separate from the main line for weeks and sometimes months before pushing to the public version control in order to minimize breakage and other issues. Take for example the planned moving of libpurple to using GObjects internally. This is a project I hope to assist with, and much of the work will likely be done privately in a local monotone database, then pushed periodically into a dedicated branch and merged for Pidgin 3.0.0 when the time is right. Between pushes, however, we have the freedom to break stuff as much as we want, then go and fix it whenever we want without having to worry about breaking things for other developers and users.
As far as plugins go, good for you that you had revived a plugin. Yes, the core crowd is a bit condescending and irritable, but realize the crap that we see in #gaim--all we ask is that people read the damn documentation and the channel topic. However, if you're making an honest development effort, we will assist you if we are able. For the most part, however, Pidgin is extremely well-documented for development, and what documentation lacks, other plugin code can often be used as an example (I have done this more times than I can count in the development of my own plugins). This abundance of documentation and examples means we expect you to do a little work for yourself, which seems to be a problem for the majority of visitors to #gaim.
For the record, I will note that I am a channel regular and have been for over three years. I am not officially affiliated with the project, but I have contributed in the past. I just happen to share some of the opinions of some of the developers and more involved contributors.
Pidgin hasn't really been surpassed in its core focus--textual instant messaging. Yes, other clients are equals in many respects. Yes, some clients have integrated that fabled voice and video support that so many users seem to want. This doesn't really mean that any application is better than Pidgin or that Pidgin has fallen behind the other clients.
A unified instant messaging standard is the point of XMPP, which is more commonly known as Jabber. It is a completely open, standards-based specification using XML, which makes it flexible and extensible. Google Talk is helping XMPP gain popularity, but to an extent hiding some of the details from its users. For widespread acceptance, at some point the details have to be hidden, and Google Talk is at least doing a decent job of it.
Dividing effort is another issue entirely. Pidgin had long wished to finish its fabled Core/UI split that started way back at Gaim 0.60 (and its nine-month GTK+2-ification process between 0.59 and 0.60), and at the 2.0.0beta4 release finally accomplished this. The few revisions in Subversion that accomplished this were a complete disaster that could have been avoided had there been a bit more patience, but what's done is done. At any rate, libpurple exists now and its purpose is to make it easy to write alternative user interfaces. Enter Finch, the ncursesw-based console UI. If everyone trying to implement voice and video in other projects could come together and get a decent abstraction layer built into libpurple, any UI that wanted to could take advantage of libpurple functionality, thus reducing duplicated effort to the frontend that the user sees, which is a significant improvement over duplicating literally everything.
Next I'd like to address paying for Pidgin. In the past this was not possible for numerous reasons, including taxing and trusting individual people with the money. Now, however, when the infrastructure is in place, anyone who wants will be able to "pay" for Pidgin by donating to the project and the Instant Messaging Freedom Corporation. Just be patient a bit longer and such things will be in place so anyone who wishes to contribute money may do so.
Let me finish by coming back to my original point--Pidgin is extremely good at what it does, and has not fallen behind.
Pidgin has never been a GNOME application, even when it was originally named Gaim. Sure, it uses GTK+, but that in itself does not make an application a GNOME application. I will grant you that an inordinate amount of focus has gone into making Pidgin play nice on GNOME with no similar effort spent on KDE or XFCE integration, but this again does not make Pidgin a GNOME app.
That project is called gaim-vv. It's experimental voice and video extensions for gaim. These are the relevant site and project page.
Development has been temporarily stopped pending some work on gstreamer. The primary developer of gaim-vv is one of the gaim developers and hopes that when he feels it's ready gaim-vv will be merged back into gaim.
Actually, that's wrong. They've been aiming for a three-week release schedule, where they like to have a release out Thursday night or Friday morning of that week. Since they went to a three-week schedule they've only been late once (with the 1.0.0 release, actually), and that was only by a few hours.
Sourceforge.net's activity statistics hasn't included CVS statistics in quite some time. Instead, activity stats are based on downloads, tracker activity (add, close, and delete bugs, feature requests, etc.), and forum activity. CVS activity is supposed to have a weight in the stats, but CVS activity hasn't been included in the stats for some time. A project's activity rating on sourceforge is not a measure of the stability or usability of the project's code. Read this, as it explains exactly what I just said.
First flaw... you're saying it's possible for/. to look good. It's easy to use once you figure it out, no question, but the designers completely ignored aesthetics (which I'm fine with.)
Well, I'm not saying/. looks beautiful. I'm just saying it looks better in some browsers than in others.
Second, more important flaw... IE, Mozilla, Opera, and Safari render/. exactly the same for me, with the only difference being that I don't see ads while using Safari. Safari handles all the pages I visit well enough to certainly never think there's a flaw in the way it's doing things.
I can tell a difference easily between IE, Mozilla, Opera, etc. Most notable are differences in color, although most people tell me I'm just imagining it. But I can also tell a difference in the way the browsers render the table code/. uses to accomplish the layout.
Rendering speed is irrelevant for me--I'm on dialup so I don't notice a rendering speed difference because it takes forever to download the page anyway.
As for saving the open tabs, I find it extremely useful as the power in my area is prone to frequent half-second outages. And the theming and HIG don't bother me in the least--I don't care how the UI looks as long as it lets me get the job done. To each his own, though.
Safari is based on Konqueror's rendering engine. While a great engine at heart, it's not as good on some sites as other rendering engines, such as Mozilla's, IE's, and Opera's, are. Granted this has some to do with the design of the site, but design isn't everything. The browser has to help make it look good, too.
Take, for example, Slashdot itself. Try viewing it in several different browsers. Everyone I know find that Opeara and IE tie for first place in making the site look good, with Mozilla/Netscape 6+ as a close second, but Konqueror as a distant third.
Opera, besides its excellent rendering engine, also has the tabbed interface working in its favor. Sure Mozilla has this too, but Opera lets you reopen the browser after a crash or application close and have all the pages that were open at the time of the crash or close. This is a lifesaver at times, for example when your cpu cooler dies and the system overheats, causing it to halt. When you repair the system and return it to operation, you can reopen Opera and have all the pages you were looking at before brought back without having to manually reopen them or hunt for them.
I'll take Opera and Mozilla over the others any day.
Any app that has autosave is already halfway there. All anyone would need to do is just keep saving each revision in the same file with a timestamp and the app would automatically find the newest timestamp and use that revision. Of course there'd have to be some sort of configuration option for when to start throwing old revisions away (say maybe after 6 days or 30 revisions or whatever, make it user customizeable).
Don't shoot me for mentioning an MS product, but if you used Office and saved everything in XML format, the revisions with timestamps could be done as simply as running it through windiff and putting a timestamp/revision number on it. It needn't be limited to MS Office though; OpenOffice, KOffice, and the like could implement and use this as well.
Moral of the story: half the battle's won; someone just needs to code the other half.
Yes, it can. It's just buried deeper than most people care to look. Unfortunately it's only an accept or drop setting; you can't configure stateful filtering.
Windows XP does tab-complete by default (totally out-of-the-box), but it sucks horribly. If you're using XP and this is not working, either someone explicitly disabled it at the OEM or (more likely) someone turned it off using TweakUI.
Actually, no. M$ only sees money for PCs that have M$ OSes on them. However, under the OEM-M$ agreements, OEMs can't sell PCs without an OS. The agreement doesn't limit the OEMs to M$ OSes; it just limits them in that all PCs shipping from OEMs having agreements with Microsoft must have any OS on/with them. Shipping a PC without an OS in the box would be breaking the contract and then open Dell and the other OEMs up to costly breach of contract suits.
Because often buying the OEM PC with XP Professional yields better hardware, such as a real video card (as opposed to the shared memory crap you commonly find in your low-end Windows XP Home Edition PCs).
I saw a demonstration by one of Panasonic's sales reps in October of 2002 (Pittsburgh ITEC event). The man dropped their then largest and most expensive Toughbook from a 6 foot height, climbed down the ladder, picked it back up, and it was still in perfect working order. I wish they weren't so featureless and expensive, though.
That's why on price sheets you often see Athlon XP 2600+TBC, Athlon XP 2600+PC, or Athlon XP 2600+BC (for Thoroughbred core, Palomino core, and Barton core, respectively). Also, on the chip itself, there is an identifier that indicates the core type. Googling for that will turn up the info needed to determine the core being used. But realistically, if all three deliver the same performance as a classic Athlon pumped to 2.6 GHz, why should it matter what core is in them (bus speed issues aside)? To me, that's just saying the Barton core is better than the Thoroughbred core because the Barton can deliver the same performance at a lower clock speed.
In the case of these Radeons, if the 9000 can deliver the exact same performance as the 9200, who cares that the chip is really a "9000"? Technically, if you look at all the different Radeons, you'll find that the chips are all labeled "Rage 6 Rxxx" where xxx is a particular revision number. In the end, it's still the same chip. And we've already learned that the 9000 and 9200 are identical other than that the 9200 is AGP8x capable. Applying this back to the Athlon XP, all three of those Athlon XPs deliver the same performance, but with different FSB speeds. Same type thing. And given that the word Mobility has been used to identify it as a Mobile Radeon 9200, it seems to me that ATI is doing nothing different from what AMD is doing.
Ah, but they do. Look closely: label advertising the Mobility Radeon 9200 brand (emphasis mine). As far as I can remember, there has ALWAYS been a difference between the modeling numbers for the mobile Radeons and the desktop Radeons, hence the reason "Mobility" was always part of the chip's name.
As far as this goes, I don't see a problem with it. Now if they were advertising it as a 4x AGP video card and it really was a PCI card, then I'd be pissed and ready to sue anyone and everyone I could.
Slashdot Mirror
There is already such a solution. It's called RFC 2369, and it dictates headers to be inserted into every message. It's existed for years; the problem is webmail and most mail client developers are too stupid or too selfish to implement support for these headers.
Evolution replaces Outlook, not Exchange.
I have just addressed this in this comment: http://linux.slashdot.org/comments.pl?sid=229897&c id=18655919
Please ignore my hostile attitude there as it is mostly directed at the anonymous post to which I was responding. In short, I am used to Pidgin's interface and don't like any other IM client's interface because of being so used to Pidgin.
- I dislike its interface, as I have grown very accustomed to Pidgin's interface after having used Gaim/Pidgin since the days of 0.43.
- I dislike the fact that when I try to combine multiple buddies into a "metacontact" in Adium, the buddy I have dragged and dropped becomes the priority buddy in the meta. This is just plain stupid. This is were Pidgin is superior, in that I can choose the buddy's priority in the contact when I create the contact instead of having to go to some stupid info dialog that should have NOTHING to do with the organization of a contact.
- I dislike the fact that the accounts and preferences dialogs are one and the same in Adium.
- I dislike the conversion of "/me does action" to "*does action*".
- I dislike the tabbed conversation windows having their tabs along the bottom and not being able to change said placement.
- And finally, I disagree on a fundamental level with the iTunes integration added to Adium. This is only an incentive for more abuse of status messages to display stupid crap like the currently playing song.
This is the first time I have ever publicly aired my complaints about Adium, and had hoped it would never happen. I will note that there are a few other reasons that I will never discuss with anyone, as a discussion of these issues will only cause more problems than it could ever solve. But, this being slashdot, the land of stupidity and flame wars, I was dragged in by a troll hook, line, and sinker and convinced to air some of my grievances. My only hope is that people are intelligent enough not to take my opinion as being that of any member of the Gaim/Pidgin development team that I am NOT officially affiliated with.I'm not fully briefed on the decision process for monotone. I believe several of the developers felt monotone to be the most intuitive for handling conflicts and merges and such, but I could be mistaken. I know several of the developers, once trying monotone, darcs and one other whose name I can't recall at the moment (perhaps git or mercurial?), preferred monotone for their own use.
I have a Mac. I use Adium for my limited OSX IM needs. I hate it with a passion.
The names actually make sense. A pidgin is basically a mish-mash of languages, which is what we see on IM with all this "u", "o rly?", "omg" and similar crappy "abbreviations" floating around. It also sounds like "pigeon", which is where the new logo will be coming in eventually. Finch was named so to keep in with the bird theme, and I suspect to imply a "little brother" idea since it is young in comparison to Pidgin itself. Libpurple comes from the abbreviation long used for protocol plugins in the client, "prpl", which is supposed to be pronounced "purple." Since libpurple's major function is to provide the IM service handling backend for messaging applications, it made sense that the protocol plugin's abbreviation should have some part in its new name.
This is not saying that the developers don't want people hacking on Pidgin. There are currently plans in place to implement a Subversion gateway so that casual hackers can pull the Pidgin source and create their patches and whatnot. The reason behind the switch to monotone is that a distributed version control system fits more in line with the core developers' workflow, working on things separate from the main line for weeks and sometimes months before pushing to the public version control in order to minimize breakage and other issues. Take for example the planned moving of libpurple to using GObjects internally. This is a project I hope to assist with, and much of the work will likely be done privately in a local monotone database, then pushed periodically into a dedicated branch and merged for Pidgin 3.0.0 when the time is right. Between pushes, however, we have the freedom to break stuff as much as we want, then go and fix it whenever we want without having to worry about breaking things for other developers and users.
As far as plugins go, good for you that you had revived a plugin. Yes, the core crowd is a bit condescending and irritable, but realize the crap that we see in #gaim--all we ask is that people read the damn documentation and the channel topic. However, if you're making an honest development effort, we will assist you if we are able. For the most part, however, Pidgin is extremely well-documented for development, and what documentation lacks, other plugin code can often be used as an example (I have done this more times than I can count in the development of my own plugins). This abundance of documentation and examples means we expect you to do a little work for yourself, which seems to be a problem for the majority of visitors to #gaim.
For the record, I will note that I am a channel regular and have been for over three years. I am not officially affiliated with the project, but I have contributed in the past. I just happen to share some of the opinions of some of the developers and more involved contributors.
Pidgin hasn't really been surpassed in its core focus--textual instant messaging. Yes, other clients are equals in many respects. Yes, some clients have integrated that fabled voice and video support that so many users seem to want. This doesn't really mean that any application is better than Pidgin or that Pidgin has fallen behind the other clients.
A unified instant messaging standard is the point of XMPP, which is more commonly known as Jabber. It is a completely open, standards-based specification using XML, which makes it flexible and extensible. Google Talk is helping XMPP gain popularity, but to an extent hiding some of the details from its users. For widespread acceptance, at some point the details have to be hidden, and Google Talk is at least doing a decent job of it.
Dividing effort is another issue entirely. Pidgin had long wished to finish its fabled Core/UI split that started way back at Gaim 0.60 (and its nine-month GTK+2-ification process between 0.59 and 0.60), and at the 2.0.0beta4 release finally accomplished this. The few revisions in Subversion that accomplished this were a complete disaster that could have been avoided had there been a bit more patience, but what's done is done. At any rate, libpurple exists now and its purpose is to make it easy to write alternative user interfaces. Enter Finch, the ncursesw-based console UI. If everyone trying to implement voice and video in other projects could come together and get a decent abstraction layer built into libpurple, any UI that wanted to could take advantage of libpurple functionality, thus reducing duplicated effort to the frontend that the user sees, which is a significant improvement over duplicating literally everything.
Next I'd like to address paying for Pidgin. In the past this was not possible for numerous reasons, including taxing and trusting individual people with the money. Now, however, when the infrastructure is in place, anyone who wants will be able to "pay" for Pidgin by donating to the project and the Instant Messaging Freedom Corporation. Just be patient a bit longer and such things will be in place so anyone who wishes to contribute money may do so.
Let me finish by coming back to my original point--Pidgin is extremely good at what it does, and has not fallen behind.
Pidgin has never been a GNOME application, even when it was originally named Gaim. Sure, it uses GTK+, but that in itself does not make an application a GNOME application. I will grant you that an inordinate amount of focus has gone into making Pidgin play nice on GNOME with no similar effort spent on KDE or XFCE integration, but this again does not make Pidgin a GNOME app.
The name is Pidgin, not Pidgin IM. Therefore PIM is inaccurate.
Development has been temporarily stopped pending some work on gstreamer. The primary developer of gaim-vv is one of the gaim developers and hopes that when he feels it's ready gaim-vv will be merged back into gaim.
Actually, that's wrong. They've been aiming for a three-week release schedule, where they like to have a release out Thursday night or Friday morning of that week. Since they went to a three-week schedule they've only been late once (with the 1.0.0 release, actually), and that was only by a few hours.
Also, it's Gaim or gaim, never GAIM.
Well, I'm not saying /. looks beautiful. I'm just saying it looks better in some browsers than in others.
I can tell a difference easily between IE, Mozilla, Opera, etc. Most notable are differences in color, although most people tell me I'm just imagining it. But I can also tell a difference in the way the browsers render the table code /. uses to accomplish the layout.
Rendering speed is irrelevant for me--I'm on dialup so I don't notice a rendering speed difference because it takes forever to download the page anyway.
As for saving the open tabs, I find it extremely useful as the power in my area is prone to frequent half-second outages. And the theming and HIG don't bother me in the least--I don't care how the UI looks as long as it lets me get the job done. To each his own, though.
Well, I never said /. was beautiful, just that it looks better in some browsers than others.
Take, for example, Slashdot itself. Try viewing it in several different browsers. Everyone I know find that Opeara and IE tie for first place in making the site look good, with Mozilla/Netscape 6+ as a close second, but Konqueror as a distant third.
Opera, besides its excellent rendering engine, also has the tabbed interface working in its favor. Sure Mozilla has this too, but Opera lets you reopen the browser after a crash or application close and have all the pages that were open at the time of the crash or close. This is a lifesaver at times, for example when your cpu cooler dies and the system overheats, causing it to halt. When you repair the system and return it to operation, you can reopen Opera and have all the pages you were looking at before brought back without having to manually reopen them or hunt for them.
I'll take Opera and Mozilla over the others any day.
Don't shoot me for mentioning an MS product, but if you used Office and saved everything in XML format, the revisions with timestamps could be done as simply as running it through windiff and putting a timestamp/revision number on it. It needn't be limited to MS Office though; OpenOffice, KOffice, and the like could implement and use this as well.
Moral of the story: half the battle's won; someone just needs to code the other half.
Yes, it can. It's just buried deeper than most people care to look. Unfortunately it's only an accept or drop setting; you can't configure stateful filtering.
Windows XP does tab-complete by default (totally out-of-the-box), but it sucks horribly. If you're using XP and this is not working, either someone explicitly disabled it at the OEM or (more likely) someone turned it off using TweakUI.
Actually, no. M$ only sees money for PCs that have M$ OSes on them. However, under the OEM-M$ agreements, OEMs can't sell PCs without an OS. The agreement doesn't limit the OEMs to M$ OSes; it just limits them in that all PCs shipping from OEMs having agreements with Microsoft must have any OS on/with them. Shipping a PC without an OS in the box would be breaking the contract and then open Dell and the other OEMs up to costly breach of contract suits.
Because often buying the OEM PC with XP Professional yields better hardware, such as a real video card (as opposed to the shared memory crap you commonly find in your low-end Windows XP Home Edition PCs).
I saw a demonstration by one of Panasonic's sales reps in October of 2002 (Pittsburgh ITEC event). The man dropped their then largest and most expensive Toughbook from a 6 foot height, climbed down the ladder, picked it back up, and it was still in perfect working order. I wish they weren't so featureless and expensive, though.
In the case of these Radeons, if the 9000 can deliver the exact same performance as the 9200, who cares that the chip is really a "9000"? Technically, if you look at all the different Radeons, you'll find that the chips are all labeled "Rage 6 Rxxx" where xxx is a particular revision number. In the end, it's still the same chip. And we've already learned that the 9000 and 9200 are identical other than that the 9200 is AGP8x capable. Applying this back to the Athlon XP, all three of those Athlon XPs deliver the same performance, but with different FSB speeds. Same type thing. And given that the word Mobility has been used to identify it as a Mobile Radeon 9200, it seems to me that ATI is doing nothing different from what AMD is doing.
As far as this goes, I don't see a problem with it. Now if they were advertising it as a 4x AGP video card and it really was a PCI card, then I'd be pissed and ready to sue anyone and everyone I could.