Honestly, how many parallel or USB cables do you need? I think I've bought a grand total of one of each type for printing purposes over the years; they are removable after all.
Try switching to the "split ebuild" kdepim-meta package instead. It pulls in the standalone kmail package as a dependency, which is what I've been successfully using for months now.
Bloat is the fact that each is designed around a different framework
Bonus points: name an OS - any OS - that only includes one application framework. Windows has MFC and.Net. Mac has Classic, Carbon, and Cocoa. Solaris had CDE and OpenWindows. Amiga had Gadtools, MUI, Reaction, etc.
Somehow, though, it all boils to "KDE and Gnome are bloated", even though every single widely-used system ever invented went through (or is currently in) the same situation.
Besides, I disagree on principle that this indicates bad design. KDE and Gnome seem to have fundamentally different approaches to several core functions. It's not fair to refer to them as redundant because they're really not once you get below the surface.
At any rate, the top poster's position was that KDE is, by itself, bloated. Your point addresses a different non-issue altogether.
People who say that Gnome or KDE are bloated need to be slapped, because they invariably have no idea what "bloat" means.
Both DEs are designed around a functional, reusable framework. In essence, every single thing you see is like a shared library. This allows the end-user applications to have a huge amount of functionality with little work, and is almost the antithesis of bloat.
If KMyMoney had code to allow me to load and save my accounts over an SFTP (or IMAP or webdav) connection, I'd agree with you. However, it simply uses the kio-slave features of KDE to support that automatically (as does almost every other KDE application). It's not bloat to include an excellent programmer's editor in every application when that editor is written as an embeddable object. It's not bloat when Konqueror can view PDFs because KPDF is written as an embeddable object.
I really don't think you have any idea what you're talking about. Writing the same code individually for each application would be a tremendous waste of resources. Designing the system from the ground up to lean heavily on reusable objects and a featureful core system is nothing but good.
I go through long dry spells, too. I think you're "rewarded" by not visiting Slashdot for a while, since it seems like I get points every time I get back from extended out-of-town trips.
That would be like taking your car into the mechanic with a transmission problem and having him scoff at you, "Pffft. Go build yourself a new one. Moron."
I'll go one further: I'm perfectly capable of hacking out HTML as the need arises, but sometimes I don't want to. I've used my.yahoo.com as my homepage for years because it's trivially easy to add new information. Want Reuters headlines? Click - it's there. Want to track mutual funds? Click - there they are. Want to track the weather both where I'm currently at and where I expect to be next week? Click - done.
There's no reason why I couldn't do that stuff myself, but Yahoo!'s put a slick interface on it and made it more convenient than I could've done on my own. My days are full enough as it is. This is one less thing to mess around with and I like it.
Amen. I just moved from the "that looks nice, but..." category to the "here's my Visa" queue. $599 now buys me the computer I want, rather than a down payment on the computer I'd use as a starting point.
Could be; the difference being that I've actually heard of and used mdmfs under FreeBSD while I haven't so much as looked for the equivalent under Linux.
Put another way, I never said or implied that Linux doesn't have the same functionality. I just said that FreeBSD does have it, and therefore the idea is interesting to me. Sorry if that's not as controversial as you seem to have hoped.:-)
FreeBSD allows you to allocate a dynamically resizable filesystem out of swap (see: md, mfs). I'm thinking of mounting the whole thing as a super-fast swap partition - basically, as a giant L4 cache - and mounting/tmp and a few other speed-critical filesystems out of there.
Mmmm, hyper-fast builds that don't depend on the latency of moving parts...
we shouldn't blame spammers for spam, we should blame the [...] IDIOTS who actually BUY from them
First, yes, we should blame spammers for spam. You blame pickpockets for stealing wallets that aren't attached to their owners with titanium chains, don't you? The fact that spammers (possibly) get responses from gullible folk doesn't lessen their responsibility.
Second, I've not seen any proof that selling products via spam is profitable. One of a friend's customers was approached by a spammer who offered to "advertise" their business. Now, whether that customer ever saw a single penny of revenue (let alone enough to offset the losses from their annoyed customers), the spammers would have made their money. In this case, at least, the spammers couldn't care less whether a single person bought the advertised product, as long as they got paid to send the emails.
Just remember, heatsink paste is an insulator, not a conductor. It just happens to be more conductive than air (which isn't saying much). You want to use the bare minimum amount of paste sufficient to avoid having air pockets between conducting surfaces, but no more than that.
I'm not saying that you don't already do this, but I've lost track of how many "more is better!" applications I've heard of spontaneously leaking magic smoke.
Mainly, I just wanted to say that there are methods other than cookies that work just as well.
I disagree. They do work, but definitely not "just as well", unless you enjoy having to add "&sessid=1234" or "<input type="hidden" name="sessid" value="1234"/>" to every single link, form, and redirect on your entire site (knowing full well that if you miss one then the session is lost).
Fortunately, I have the privilege of maintaining a member's-only site where we can tell customers that they have to enable cookies in order to use it. I feel bad for others who have to cater to the paranoid by shifting the session id from a small text file that gets forgotten at the end of the visit to an entity in the URL that gets forgotten at the end of the visit (given that we're only talking about session cookies here).
Given that the two are exactly functionally identical, I have little sympathy for those whining visitors. I'd wager that most of them heard "cookies are t3h 3v1l!!1!" and believe it without really understand what it meant.
Cashier: May I have your phone number?
Me: Sure! It's $(friend's ex-wife's phone number), and I'll love to hear more about other promotions you may have in the future.
"Funny" can cost someone karma since they don't get points for it, but lose some if the post is later modded "Overrated" or similar. For example, if I got modded up "Funny" three times and then back down to the original score via "Redundant", then I lose three points.
Because of this, a lot of mods now avoid the "Funny penalty" by using different moderation categories. Besides, maybe someone thought the point behind my sarcasm was indeed insightful?
When they get out into the real world, what practical knowledge will their experience with Linux afford them?
Preach it, brother. I have been endlessly cursed by an early exposure to a Timex ZX-81 and Commodore 64, and may never recover from once having owned an Amiga. Oh, my kingdom for the ability to somehow acquire new skills that are similar to the ones I already have!
No, I want my kids learning XP and only XP, and that's been my opinion ever since the United Nations declared it the One True OS For Posterity. I don't want them to look back with shame and horror on the weird systems of their youth as they attempt to learn the Windows path 30 years from now (which will be exactly identical in to current systems - how could we ask them to cope with change?).
I would not necessarily trust Snopes in all instances either.
Likewise. I read their dissection of the rumor that Marilyn Monroe had six toes, and among their "evidence" was the idea that it would've taken her a long time to recover from the amputation of the extra digit and to relearn to walk.
I asked my wife, a podiatric surgeon, if that was true. Her response was that she lopped off pinky toes all the time (she sees lots of diabetics with circulation problems) and that the healing mainly consists of the incision closing up correctly.
I wrote to Snopes to tell them that they might want to reconsider this particular bit of "proof". Their response is that everyone knows you can't walk correctly without all your toes. WTF?!? I have much more faith in the opinion of a medical professional specializing in that exact body part than "common knowledge", and I'd certainly hope that Snopes (of all groups) would feel the same way. Seems I was wrong.
At least they did choose to believe me when I also pointed out (regarding another page on their site) at aspirin and caffeine do indeed have a synergistic effect, despite what they'd previously said on the subject. Apparently the editors do occasionally accept feedback.
I guess that what's true of the news is also true of Snopes: it's almost always accurate, except when you happen to know something about the subject at hand.
Re:Of course it isn't dead!
on
DECnet Isn't Dead
·
· Score: 2, Informative
Who the heck CARES how many "Moore" generations a computer is if it does the job it was designed to do?
The point of that was that new hardware is so much faster than anything made in the '70s that you can get massive redundancy and a huge performance increase - there's not really a tradeoff. Suppose that one current server (I never said anything about desktops) can outperform a whole VAX cluster, which is very likely to be true. Put 10 of them in a redundant parallel system and I'm sure you can get the same reliability as the VAX setup.
The network doesn't. The network admins, though, tend not to be overly concerned with opening ports for P2P apps (regardless of what you plan on using it for).
keep in mind that active FTP has, and continues to work, just fine for thousands of users, for like 20 years now
Certainly, but it's dead and buried for millions more.
Re:Of course it isn't dead!
on
DECnet Isn't Dead
·
· Score: 2, Insightful
Indeed, it is by far the most superior form of networking out there for applications where the uptime must be literally 100%.
What properties make it better suited than TCP/IP for critical stability? I admit that I don't know much about many of the "old" protocols, so I probably lack the perspective to really appreciate one over another (kind of like functional programming seems goofy until you try it - then enlightenment!).
Put another way, why is VAC/Alpha-over-DECnet better than VAC/Alpha-over-TCP/IP?
Re:Of course it isn't dead!
on
DECnet Isn't Dead
·
· Score: 0, Flamebait
I'd trust my 25 year old VAX cluster over any PC-based system or cluster any day.
I wouldn't. You're running hardware nearly 17 "Moore generations" old; a similarly sized cluster of decent new servers should get you in the neighborhood of 100,000 times the throughput. If you can get a day's worth of processing on the old system done in less than a second, you can dedicate the new cluster to reliability (say, 10x redundancy and only 10,000 times the performance) and still come out with a clearly better system.
I love old tech, but I don't think you can decisively say it's "better". I have no doubt that it has huge error tolerances, but there are other ways to get that sort of robustness now.
Slashdot Mirror
Honestly, how many parallel or USB cables do you need? I think I've bought a grand total of one of each type for printing purposes over the years; they are removable after all.
That other stuff? That's still pretty evil.
Try switching to the "split ebuild" kdepim-meta package instead. It pulls in the standalone kmail package as a dependency, which is what I've been successfully using for months now.
Bonus points: name an OS - any OS - that only includes one application framework. Windows has MFC and .Net. Mac has Classic, Carbon, and Cocoa. Solaris had CDE and OpenWindows. Amiga had Gadtools, MUI, Reaction, etc.
Somehow, though, it all boils to "KDE and Gnome are bloated", even though every single widely-used system ever invented went through (or is currently in) the same situation.
Besides, I disagree on principle that this indicates bad design. KDE and Gnome seem to have fundamentally different approaches to several core functions. It's not fair to refer to them as redundant because they're really not once you get below the surface.
At any rate, the top poster's position was that KDE is, by itself, bloated. Your point addresses a different non-issue altogether.
People who say that Gnome or KDE are bloated need to be slapped, because they invariably have no idea what "bloat" means.
Both DEs are designed around a functional, reusable framework. In essence, every single thing you see is like a shared library. This allows the end-user applications to have a huge amount of functionality with little work, and is almost the antithesis of bloat.
If KMyMoney had code to allow me to load and save my accounts over an SFTP (or IMAP or webdav) connection, I'd agree with you. However, it simply uses the kio-slave features of KDE to support that automatically (as does almost every other KDE application). It's not bloat to include an excellent programmer's editor in every application when that editor is written as an embeddable object. It's not bloat when Konqueror can view PDFs because KPDF is written as an embeddable object.
I really don't think you have any idea what you're talking about. Writing the same code individually for each application would be a tremendous waste of resources. Designing the system from the ground up to lean heavily on reusable objects and a featureful core system is nothing but good.
I go through long dry spells, too. I think you're "rewarded" by not visiting Slashdot for a while, since it seems like I get points every time I get back from extended out-of-town trips.
I'll go one further: I'm perfectly capable of hacking out HTML as the need arises, but sometimes I don't want to. I've used my.yahoo.com as my homepage for years because it's trivially easy to add new information. Want Reuters headlines? Click - it's there. Want to track mutual funds? Click - there they are. Want to track the weather both where I'm currently at and where I expect to be next week? Click - done.
There's no reason why I couldn't do that stuff myself, but Yahoo!'s put a slick interface on it and made it more convenient than I could've done on my own. My days are full enough as it is. This is one less thing to mess around with and I like it.
That begs the question: what do we get?
Or 560,000x if calculated on a PowerPC with AccurateIntegerMath (tm) technology.
Amen. I just moved from the "that looks nice, but..." category to the "here's my Visa" queue. $599 now buys me the computer I want, rather than a down payment on the computer I'd use as a starting point.
Could be; the difference being that I've actually heard of and used mdmfs under FreeBSD while I haven't so much as looked for the equivalent under Linux.
Put another way, I never said or implied that Linux doesn't have the same functionality. I just said that FreeBSD does have it, and therefore the idea is interesting to me. Sorry if that's not as controversial as you seem to have hoped. :-)
I don't have a motherboard that can hold 4GB of RAM, so I'd tend to disagree.
Yeah, I think I might have to snag a couple of these.
Mmmm, hyper-fast builds that don't depend on the latency of moving parts...
First, yes, we should blame spammers for spam. You blame pickpockets for stealing wallets that aren't attached to their owners with titanium chains, don't you? The fact that spammers (possibly) get responses from gullible folk doesn't lessen their responsibility.
Second, I've not seen any proof that selling products via spam is profitable. One of a friend's customers was approached by a spammer who offered to "advertise" their business. Now, whether that customer ever saw a single penny of revenue (let alone enough to offset the losses from their annoyed customers), the spammers would have made their money. In this case, at least, the spammers couldn't care less whether a single person bought the advertised product, as long as they got paid to send the emails.
Just remember, heatsink paste is an insulator, not a conductor. It just happens to be more conductive than air (which isn't saying much). You want to use the bare minimum amount of paste sufficient to avoid having air pockets between conducting surfaces, but no more than that.
I'm not saying that you don't already do this, but I've lost track of how many "more is better!" applications I've heard of spontaneously leaking magic smoke.
Found it on the bathroom wall. It's common knowledge in these parts.
I disagree. They do work, but definitely not "just as well", unless you enjoy having to add "&sessid=1234" or "<input type="hidden" name="sessid" value="1234" />" to every single link, form, and redirect on your entire site (knowing full well that if you miss one then the session is lost).
Fortunately, I have the privilege of maintaining a member's-only site where we can tell customers that they have to enable cookies in order to use it. I feel bad for others who have to cater to the paranoid by shifting the session id from a small text file that gets forgotten at the end of the visit to an entity in the URL that gets forgotten at the end of the visit (given that we're only talking about session cookies here).
Given that the two are exactly functionally identical, I have little sympathy for those whining visitors. I'd wager that most of them heard "cookies are t3h 3v1l!!1!" and believe it without really understand what it meant.
Here's how I handle it:
Cashier: May I have your phone number?
Me: Sure! It's $(friend's ex-wife's phone number), and I'll love to hear more about other promotions you may have in the future.
Because of this, a lot of mods now avoid the "Funny penalty" by using different moderation categories. Besides, maybe someone thought the point behind my sarcasm was indeed insightful?
Preach it, brother. I have been endlessly cursed by an early exposure to a Timex ZX-81 and Commodore 64, and may never recover from once having owned an Amiga. Oh, my kingdom for the ability to somehow acquire new skills that are similar to the ones I already have!
No, I want my kids learning XP and only XP, and that's been my opinion ever since the United Nations declared it the One True OS For Posterity. I don't want them to look back with shame and horror on the weird systems of their youth as they attempt to learn the Windows path 30 years from now (which will be exactly identical in to current systems - how could we ask them to cope with change?).
Likewise. I read their dissection of the rumor that Marilyn Monroe had six toes, and among their "evidence" was the idea that it would've taken her a long time to recover from the amputation of the extra digit and to relearn to walk.
I asked my wife, a podiatric surgeon, if that was true. Her response was that she lopped off pinky toes all the time (she sees lots of diabetics with circulation problems) and that the healing mainly consists of the incision closing up correctly.
I wrote to Snopes to tell them that they might want to reconsider this particular bit of "proof". Their response is that everyone knows you can't walk correctly without all your toes. WTF?!? I have much more faith in the opinion of a medical professional specializing in that exact body part than "common knowledge", and I'd certainly hope that Snopes (of all groups) would feel the same way. Seems I was wrong.
At least they did choose to believe me when I also pointed out (regarding another page on their site) at aspirin and caffeine do indeed have a synergistic effect, despite what they'd previously said on the subject. Apparently the editors do occasionally accept feedback.
I guess that what's true of the news is also true of Snopes: it's almost always accurate, except when you happen to know something about the subject at hand.
The point of that was that new hardware is so much faster than anything made in the '70s that you can get massive redundancy and a huge performance increase - there's not really a tradeoff. Suppose that one current server (I never said anything about desktops) can outperform a whole VAX cluster, which is very likely to be true. Put 10 of them in a redundant parallel system and I'm sure you can get the same reliability as the VAX setup.
The network doesn't. The network admins, though, tend not to be overly concerned with opening ports for P2P apps (regardless of what you plan on using it for).
keep in mind that active FTP has, and continues to work, just fine for thousands of users, for like 20 years now
Certainly, but it's dead and buried for millions more.
What properties make it better suited than TCP/IP for critical stability? I admit that I don't know much about many of the "old" protocols, so I probably lack the perspective to really appreciate one over another (kind of like functional programming seems goofy until you try it - then enlightenment!).
Put another way, why is VAC/Alpha-over-DECnet better than VAC/Alpha-over-TCP/IP?
I wouldn't. You're running hardware nearly 17 "Moore generations" old; a similarly sized cluster of decent new servers should get you in the neighborhood of 100,000 times the throughput. If you can get a day's worth of processing on the old system done in less than a second, you can dedicate the new cluster to reliability (say, 10x redundancy and only 10,000 times the performance) and still come out with a clearly better system.
I love old tech, but I don't think you can decisively say it's "better". I have no doubt that it has huge error tolerances, but there are other ways to get that sort of robustness now.