But would they know the right solution from the gazillions of incorrect (but superficially sound) alternatives? No part of the "1 million monkeys" theory takes that into account.
One thing that should definitely be covered is the 3-way "Save file" dialog that comes up when exiting a program/shutting down, and similar dialogs, that offer "Yes", "No", and "Cancel".
Which reminds me of #2 in the list of Ten Most Wanted Design Bugs, which proposes to "[c]onvert your existing software and write new software to perform Continuous Save, so users cannot lose more than the last few characters typed or gestures entered." Even better, include some Dead Simple Versioning, where the user creates some kind of "milestone" in the history to be able to revert any time. As a bonus, create an automatic milestone every time the application quits. No more saves!
This is maybe the single most important feature of Wikipedia: It is simply a great starting point, and should (as all information) be taken with a grain of salt. To "fix" that you'd have to change nothing less than human nature.
That's it then, for the feature most loved about del.icio.us: The dead simple interface. Now it'll be 2+ links away from the main page, accessible only through their enormously bloated I-wish-I-was-using-lynx-right-now interface, with a password form that for some stupid reason doesn't work with the Password Generator.
I guess I'll just have to backup my 2393 bookmarks and 3482 unique tags (as of today) and, dunno, hope someone builds an open source version.
OK, since we're in flame mode, I'll try to give you a bit clearer explanation of the situation. First of all, I'd love to see open source support for my hardware. In fact so much that I'm planning to buy only hardware with proper driver support on F/OSS OSes. Oh, and I was not talking about myself regarding the MSc, Xen hacking, or using Gentoo. I have a long way to go:-) But it was me who didn't find a non-broken Tulip driver before a friend helped.
When I can do better on $1 than you can on $100 maybe it's time you accept that there's something wrong with you.
Good one. Maybe you'd like to have a look at my open source projects one day? I could give you the web address and CVSROOT.
Do you know how long the Tulip driver has been in the kernel?
Do you know how many network drivers are in the kernel? I'm not against that, it would just make setup 50 times easier if they had an entry "3Com this'n'that'" instead of just "Tulip". No newbie is interested in the name of the driver, just which one will work on his/her hardware. Anyway, this was... at least three years ago, and thank **** most distros are child's play to install now if you have compatible hardware.
Why do you people even bother arguing with me anymore?
Maybe because your post is not in touch with reality? You cannot "give them 10 out of 10 with fewer bugs" today, and it will be a long time before you can. I love open source, but there's no way in hell I'd recommend Linux to my parents.
If it works on even one distro it can work on any distro.
That's not the same as saying all hardware will work on all distros, as your original post said. Sad, but true.
[...] It will change as a larger percentage of users clue up.
You don't consider someone who has an MSc in CS, works full time porting Xen to 64 bit processors, and uses Gentoo for his work system "clued up"?! Or how about the other guy, who was administering several company Linux and Windows servers while still working on his MSc? What planet are you posting from?
Don't blame Linux if you lack Google-fu.
In this particular case, I found a driver on 3Com's pages which was so broken that I had to rename the files before make could even begin to complain properly, because the case didn't match! So don't try to tell me all drivers work out of the box.
I can give them 10 out of 10 with fewer bugs [...]
Heard it a million times, but that still doesn't make it true. 'specially considering the several weeks I've used in total on different distros to make my hardware work. Often, long time Linux users have given up after trying for hours. You may consider me stupid, but I'd rather have >95% of my hardware working five seconds after I plug it in, than using days to dig forth the simple fact that the Tulip driver is what I need for my 3Com NC. Obvious, innit?!? I could go on for hours, but the simple fact remains that users are not willing to spend neither hours nor days to get their newest USB thingie working.
Wake me up when they open-source that piece of turd syncing software, PC Suite. Or does the newer versions actually give me the opportunity to select, for each conflicting entry, which one is the newest? That, and a bit of stability and the ability to set proper defaults based on the phone model would get me right back on.
Pardon me all the way, but I'd call BS on anyone trying to convince me that s/he's build a "large scale app" without even knowing how to spell the language s/he used. No sirree.
As long as I've been conscious about the whole shittyness of business, I've not once bought a product based on advertisements. I believe that advertisements, in the very general case, is made for at least one of the following reasons:
The product is absolutely no different from others (e.g., most bathroom equipment)
The product sucks donkey balls (e.g., TV Shop products, rip-offs like X Cola)
The product idea is good, but the result has one enormously negative feature (e.g., shape, weight, fragility, price, smell, taste, usability)
For any useful thing selling at above ~$100, you can use the 'net to find dozens of reviews. Combined with a bit of common sense, those should be plenty to find out what you should buy.
Every time you hear someone say "I shouldn't have to read the manual to figure out how to use it!", you're seeing another example of the problem.
Ever wonder why Gmail is more popular than Mutt? Maybe it's because not everyone is willing to learn Vim and read two 50 page manuals to get their emails. People have different views on how to best spend their time. Obviously, me even having to say this means I don't value my time enough.
Here goes my karma, but think about these points before modding:
Service outages? When was the last time Google was down? Gmail? Slashdot? Your online bank? I'd bet the average home user has a much better image of the stability of the web than that of her own computer...
Platform issues? How about the millions of webpages which look, feel, and work the same in basically any web browser, even textual ones? Sure, you don't get "pixel perfection" all the time, but when did you last worry about that "thin" border being 2px in your less-than-favorite browser?
Version control? Ever heard about XML? ODS? I generally expect more of web services than programs, if only for the simple reason that there is actual competition out there.
Licenses? Well, how about the thousands of services which are free to use, but still make money? This isn't 1996, friend.
Slashdot Mirror
But would they know the right solution from the gazillions of incorrect (but superficially sound) alternatives? No part of the "1 million monkeys" theory takes that into account.
Reminds me of "The Adventures of ACTION ITEM(TM)"!
So we've got "ESA to Send Spacecraft to Venus" today, and "ESA Spacecraft Reaches Venus" tomorrow? Non-RTFA-ers should be mighty impressed...
Butbutbut, it's got a new name! That's got to count for something!
No - They're you!
Zen lessons, courtesy of Microsoft
In Norway, that is referred to as the harddisk. Arg.
Which reminds me of #2 in the list of Ten Most Wanted Design Bugs, which proposes to "[c]onvert your existing software and write new software to perform Continuous Save, so users cannot lose more than the last few characters typed or gestures entered." Even better, include some Dead Simple Versioning, where the user creates some kind of "milestone" in the history to be able to revert any time. As a bonus, create an automatic milestone every time the application quits. No more saves!
Too bad they haven't fixed the really annoying blocking bugs, such as not saving drafts on IMAP...
Are they still going to teach us "the intricate workings of a belt buckle"?
This is maybe the single most important feature of Wikipedia: It is simply a great starting point, and should (as all information) be taken with a grain of salt. To "fix" that you'd have to change nothing less than human nature.
That's it then, for the feature most loved about del.icio.us: The dead simple interface. Now it'll be 2+ links away from the main page, accessible only through their enormously bloated I-wish-I-was-using-lynx-right-now interface, with a password form that for some stupid reason doesn't work with the Password Generator.
I guess I'll just have to backup my 2393 bookmarks and 3482 unique tags (as of today) and, dunno, hope someone builds an open source version.
Ditto on Athlon XP 1600+ with 1 GB RAM.
The answer which comes to mind: This was done to make sure the rootkit was introduced with a low profile, to alpha-test it...
Candidate for "Unbreakable II, the reality TV series", anyone?
Three Companies Shutdown For Spyware Bundling, why shouldn't Sony have to suffer the same fate?
OK, since we're in flame mode, I'll try to give you a bit clearer explanation of the situation. First of all, I'd love to see open source support for my hardware. In fact so much that I'm planning to buy only hardware with proper driver support on F/OSS OSes. Oh, and I was not talking about myself regarding the MSc, Xen hacking, or using Gentoo. I have a long way to go :-) But it was me who didn't find a non-broken Tulip driver before a friend helped.
Good one. Maybe you'd like to have a look at my open source projects one day? I could give you the web address and CVSROOT.
Do you know how many network drivers are in the kernel? I'm not against that, it would just make setup 50 times easier if they had an entry "3Com this'n'that'" instead of just "Tulip". No newbie is interested in the name of the driver, just which one will work on his/her hardware. Anyway, this was ... at least three years ago, and thank **** most distros are child's play to install now if you have compatible hardware.
Maybe because your post is not in touch with reality? You cannot "give them 10 out of 10 with fewer bugs" today, and it will be a long time before you can. I love open source, but there's no way in hell I'd recommend Linux to my parents.
That's not the same as saying all hardware will work on all distros, as your original post said. Sad, but true.
You don't consider someone who has an MSc in CS, works full time porting Xen to 64 bit processors, and uses Gentoo for his work system "clued up"?! Or how about the other guy, who was administering several company Linux and Windows servers while still working on his MSc? What planet are you posting from?
In this particular case, I found a driver on 3Com's pages which was so broken that I had to rename the files before make could even begin to complain properly, because the case didn't match! So don't try to tell me all drivers work out of the box.
Heard it a million times, but that still doesn't make it true. 'specially considering the several weeks I've used in total on different distros to make my hardware work. Often, long time Linux users have given up after trying for hours. You may consider me stupid, but I'd rather have >95% of my hardware working five seconds after I plug it in, than using days to dig forth the simple fact that the Tulip driver is what I need for my 3Com NC. Obvious, innit?!? I could go on for hours, but the simple fact remains that users are not willing to spend neither hours nor days to get their newest USB thingie working.
Wake me up when they open-source that piece of turd syncing software, PC Suite. Or does the newer versions actually give me the opportunity to select, for each conflicting entry, which one is the newest? That, and a bit of stability and the ability to set proper defaults based on the phone model would get me right back on.
Pardon me all the way, but I'd call BS on anyone trying to convince me that s/he's build a "large scale app" without even knowing how to spell the language s/he used. No sirree.
As long as I've been conscious about the whole shittyness of business, I've not once bought a product based on advertisements. I believe that advertisements, in the very general case, is made for at least one of the following reasons:
For any useful thing selling at above ~$100, you can use the 'net to find dozens of reviews. Combined with a bit of common sense, those should be plenty to find out what you should buy.
Ever wonder why Gmail is more popular than Mutt? Maybe it's because not everyone is willing to learn Vim and read two 50 page manuals to get their emails. People have different views on how to best spend their time. Obviously, me even having to say this means I don't value my time enough.
Here goes my karma, but think about these points before modding:
Service outages? When was the last time Google was down? Gmail? Slashdot? Your online bank? I'd bet the average home user has a much better image of the stability of the web than that of her own computer...
Platform issues? How about the millions of webpages which look, feel, and work the same in basically any web browser, even textual ones? Sure, you don't get "pixel perfection" all the time, but when did you last worry about that "thin" border being 2px in your less-than-favorite browser?
Version control? Ever heard about XML? ODS? I generally expect more of web services than programs, if only for the simple reason that there is actual competition out there.
Licenses? Well, how about the thousands of services which are free to use, but still make money? This isn't 1996, friend.