You can right-click where it doesn't have any programs or on the edge, and there's a rectangle you can click+hold and drag to change size I think.
Let me put it this way: Once I unlock the widgets, I can open a "configuration" system. In there are about five or ten controls with absolutely no means of labeling them -- not even tooltips.
Ah, I see it. So I made my panel smaller, and almost everything works... Except the clock is broken. It was always broken, but at least I could read the time before. Now the time is chopped off on the top, and the date is chopped off left/right.
WTF? Did anyone even test this?
2.0, to be blunt, stunk, and it really turned me off to KDE4 since 1.4 won't start due to different audio architectures.
See, I have 1.4 working just fine, except for the bug I mentioned. The WONTFIX because you WILL upgrade to Amarok2, which is missing half the features.
And there are a lot of things I really like that were done really well.
I'll agree there. Personally, the most obvious thing I'd miss is alt+f2 (it was too easy to make katapult screw up); okular (kpdf scrolls by some amount less than a page, can't rotate the display, etc); and Konqueror4 just works with a lot more pages (Pandora, for example).
In fact, that alone is a big deal -- that plus Flash 10 means that, for example, I can just fire up a Konqueror window on Pandora and leave it running. On Hardy, I'd have to use Firefox, which means if Flash screws up in some other Firefox window, or if Firefox crashes, the music goes away.
Not sure what you mean about saving a file or browsing porn... though one other feature I kind of miss is the ability to click+drag a link to a konsole, and have it paste a kfmclient command, not just the URL. That way, it goes into KDE's cache, and gives me a nice GUI progress bar. As it is, I have to choose between pasting it there and typing 'wget', or navigating back to that folder via the GUI -- lame.
as of 4.2beta2 they've got about 90+% of "it" nailed as well as or better than 3.5.
Unfortunately, the other 10% are things people really need working.
They've somehow done the 80/20 rule, backwards.
I truly think that most of these shenanigans could have been avoided if they'd tested the final RC on 100 people who'd never used the alphas or betas
I think a single person might have made a difference.
The problem is, this is a desktop environment. Sure, I could've stayed, I could've assumed it was a bug in documentation or UI if I couldn't figure something out (and not just my own stupidity). Or I could've spent my time actually using my computer, in a mostly-working fashion, for things like working, you know, for money.
But the real question here is: Did no one on the Amarok core team actually use it for long enough to figure out how to make a fucking pause button work? Did no one ever actually use the clock applet in the panel?
That's the thing, in Intrepid you can't chose KDE 3.5
I could grab a PPA, or I could grab Hardy. Or I could use Intrepid and KDE4, but fall back to KDE3 for half the packages -- Amarok2 isn't done, so Amarok will be using the KDE3 libs, as will networkmanager (thus breaking kwallet integration)...
it isn't that far behind and it has some added interesting features.
And yet, the fact that it is behind is a big slap in the face. I've told people not to upgrade to KDE4, and I'm about to start telling them to go back to Windows. It's that bad.
I mean, really. Git's good -- git's even really good -- but what does it matter if Perl 6.x is gonna take longer than Duke Nukem Forever to come out?
Maybe Git will help it come out sooner. I know I'd much sooner contribute if I could just fork the project on Github, as opposed to sending patches against a Perforce repository.
And be careful with the DNF jokes. We said the same about Vista, and look what happened.
I'm not a big fan of ESR, but I have to admit "release early and release often" is something I happen to agree with.
Then you obviously missed that point. "Release early and often" is about making the code available, and there are beta/prereleases available for Perl6, so they've got that covered.
It absolutely should not apply to final versions. Those should be released when it's done, not before. For a good example of what happens when people release something just to release it, look at KDE4.
Git is built of a patchwork of C and scripts, meaning it's something Perl6 could be a part of someday, and it's also something that's going to be quite familiar to all Perl developers, not just the Pugs guys.
And, Git seems to be quickly becoming the Subversion of DVCS -- fast, open source, everyone has it, everyone knows it, and the alternatives really don't have much compelling to offer.
There are significant advantages of Git over Subversion. RTFS for some.
Just to add insult to injury -- often, a Git checkout, which includes all history, takes up less space than a Subversion checkout for the same project, which doesn't even include recent commit log messages.
But think about this -- you're saying they should use a big, slow, central server, as a single point of failure, crippling offline development, complicating branches (especially merges), and several orders of magnitude slower for just about every operation, just so you don't have to learn a "weird" tool?
See, and I thought it might have something to do with separating work and personal correspondence. Or maybe with branding -- every time you email someone on behalf of the company, you're both advertising the company, and providing some amount of assurance that you work for them.
Or maybe to manage mailing lists, or group calendars. Or to provide email and webmail to people who don't already have them.
The main thing is you can take regular backups of your exchange server. It is not so easy to backup GMail.
Aside from the fact that Google presumably keeps backups, and you're generally not deleting email in the first place, I don't see how it's easier to back up exchange server than Gmail.
I should qualify it, of course -- "different" sometimes is quite a lot better, and more intuitive, even revolutionary. I love KDE4's new alt+f2.
However, a lot of this just feels different for the sake of being different. I don't even mind the lack of files on the desktop -- best example is the panel configuration.
Disclaimer: I am using KDE4. I like it for what it could be. As it is, I'm looking at alternatives.
Replace "4.0" with "Vista", "4.1" with "Vista SP1", and "4.2" with "Vista SP2" -- and, for good measure, "3.5" with "XP Pro", and you have a fair sense of what's going on here.
In fact, Microsoft has handled this better -- they still fix bugs in XP.
In KDE4, and in some of the bigger KDE4 apps (like AmaroK), there's this completely new, exciting, amazing version which almost has all the features you needed from the old version, in a very cool-looking but annoyingly different way, and sometimes crashes. Then there's the old, boring, unsupported version, which does everything you want it to do, but has some annoying bugs and deficiencies -- yet whenever you point them out, people close the bug "wontfix" as development has stopped on that branch, and the KDE4 version will be done so differently the bug is irrelevant.
At least Windows has a mostly-working version -- XP. KDE has no working version.
An example of something that worked in 3, but is broken in 4: The panel. Everyone always said, "Don't mind that, it's fixed in 4.1." Well, I'm running 4.1, and I can tell you, it's not even close. How do I make the panel thinner vertically? How do I adjust its translucency -- how do I give it a completely transparent background, but solid foreground?
An example of something that doesn't work anywhere (wontfix in 3, not done yet in 4) is encoding scripts in AmaroK. There's no longer a GUI option to tell AmaroK what your preferred format for a device is -- if you've got an iPod, it's going to give you mp3s, whether you want them or not, even if you can handle AAC just fine. Yet the KDE4 version of AmaroK doesn't yet support encoding scripts in any way, so my choice is mp3s, or no encoding at all. WTF?
Maybe I'm just using the wrong distro? I was pretty appalled at Kubuntu's handling of Intrepid. Bluetooth is broken, due to conflicting versions of a few packages. The only available solutions are, use the commandline (I tried, didn't work), go back to Hardy, or use the Gnome bluetooth GUI.
Isn't that why you use a distro in the first place? So bullshit like this doesn't happen?
Here's hoping by 4.5, they'll finally attain the functionality of 3.5. Maybe they'll still have some users left by then. Meanwhile, I'm going to take a long, hard look at going back to Fluxbox or straight Compiz.
Consider this: every time the Rails team changes things, you'll have to go through your patches and make sure they still apply correctly. And if, heavens forbid, they do some major refactoring, you'll have to spend the time figuring out what functionality got moved where, and re-apply the patches as necessary.
So write plugins, instead of monkeypatching.
Better yet, use a framework like Merb, which has a well-specified Plugin API. Stick with that, and your plugin will work for at least the rest of Merb 1.x. If there's something you can't do with only the plugin API, that is considered a bug.
Merb 2.0 will also be Rails 3.0, so alternatively, wait until then -- Rails will then have a plugin API also.
Of course, caching and setting up distributed workload processing queues will be necessary in any framework, but that's up to the development team working on a specific app rather than the framework itself.
True -- but Rails does make caching easy.
You may want to look at how Rails 3.0 (the next release of the Rails framework) is incorporating several ideas from Merb to make the framework lighter-weight and more extensible
It's more than that -- Rails 3.0 is also Merb 2.0. So, Rails 3.0 should gain all of the advantages of Merb -- one of which is raw performance.
For offline processing in a Rails app, look at plugins like workling/starling and other message queues.
For what it's worth, Merb has a simple threaded backgrounder built in.
Also, kindly remind them that "different" isn't always "better", or "more intuitive", or "revolutionary". Sometimes it's just annoyingly unfamiliar, for no good reason.
A mail client that uses an open source calendar source, and integrates it well with email, would be a great Outlook replacement, but I've not seen this either.
Have you tried Kontact?
Last I checked, Kmail couldn't quite handle folders as large as Thunderbird. Other than that, there's Thunderbird with Sunbird.
Is Gmail and Google Calendar working well?
Generally, yes.
Even if it is, it raises the problem that your calendar and email are off-site and you are vulnerable to data theft and loss of services if your external connection is cut.
Haven't had the data theft happen. And unless you're talking about Google stealing data themselves, it seems no more likely than the same thing happening with your own groupware solution, assuming you allow remote access.
I suppose, technically, I could simply connect via IMAP (with a large cache, and offline browsing enabled) and ical+dav. In practice, I've been using Google Apps at work for over a year, and I've never had it go down -- and the times our connection has died, I can count on one hand. But then, it's also less of a concern -- I know I don't get much work done without Internet.
it's still a web client and still slower to work then Outlook.
Citation needed. I've had problems with it, but slowness was never one of them.
Second, many companies are not willing to turn over their email to outside party that they cannot control what they do with it.
My assumption was, there are very specific things they're allowed to do with it.
Besides which, the alternative is trusting Microsoft to not patch Outlook to send them copies of your inbox. If you don't suspect Microsoft of doing that, why do you suspect Google of doing something similar?
Zimbra is a nightmare
That's one of four that I was able to find. Haven't used any, though.
there is no reason to use Outlook as client but not use Exchange as backend.
The reason is, you want it to be a seamless upgrade. For example, replacing the Windows fileserver with Samba on Linux, done right, no one should notice except the admins. Similarly, adding webmail to an IMAP server, even if you have to change the underlying server quite a lot, is something the users shouldn't notice, unless they happen to like webmail.
So, if you have a legitimate reason to want to use something other than Exchange as a backend -- maybe better Linux compatibility, maybe it's easier to admin, maybe it has features Exchange doesn't -- it would be nice if the people using Outlook don't have to care. That's why you want to support Outlook as a client.
and why exactly would a linux admin of equal skill get more done? greater l33tness factor or something?
This is always the counterargument, but it is consistently true. I'm guessing it's got to do with Linux being easier to admin, particularly easier to script, but I don't know a lot about Windows admin.
windows typically utilises standard gui management tools. linux admin's typically utilise cryptic scripts that take even experienced admin's an hours to decypher.
That might be part of the above point on efficiency, then. If you're not scripting repetitive admin tasks, you're Doing It Wrong. And by doing it right, you free up some time to actually document what you do -- which is important no matter how many GUI tools you use.
Now, if you are scripting things, there's not really a difference here -- on Linux, you've got standard commandline management tools; on Windows, they're standard GUI management tools. Do you script that GUI? If so, I'll bet your scripts are even more cryptic and fragile than the worst Linux admin script. If not, you're wasting a lot of time on repetitive bullshit you shouldn't have to do.
well, i can pick up the phone and have MS tell me exactly what my volume license costs me and what i get with it. can you do the same with linux?
No, MS doesn't do Linux.
But seriously, yes, I can call Canonical, or I can call Redhat, or even Novell. Or, I can do the support in-house, if needed. Or I can hire a contractor and get vague estimates.
Point is, there's a choice. If you're doing Windows, you pick up the phone and call MS, full stop. If there's something the OS doesn't do that you want it to do (even a simple security hole), you call MS and grovel, or you pay through the nose for "Shared Source" and someone to code it, assuming the parts you want are even available that way.
why is it moronic? if you lose network connectivity what do you intend on doing? serial terminal?
Why not? Plug it into a neighboring machine, and hook that up to the network. Much faster than doing the same with a capture card of some sort.
If it's unbootable, pull the hard drive, put it in another machine.
you may as well have a video card since you can't buy a mobo without one these days.
Doesn't have to be bare metal -- what about a virtual machine?
Supposing it is bare metal, you apparently weren'tlooking. Why should I have to buy a video card on top of that?
Why is it any different to "requiring" a serial port or a Lights-out-management card ?
Mostly because a serial terminal can be managed remotely, and because that video card takes up more space and electricity than a serial port.
And yet, I seem to remember not needing a serial port either. It's a headless machine; let it be headless.
Incidentally, I doubt the vast majority of hardware engineers are "moronic", yet for some reason they think including a video card is a reasonably good idea.
Must be a recent phenomenon, then. I certainly remember buying motherboards without video cards, which would run fine without them. If it was unbootable for some reason, you'd go find a spare PCI video card and use that.
And yet, we're not necessarily talking about hardware, either. Consider virtual machines, and the problem compounds. You now need a virtual display somewhere, either by virtualizing access to the actual video hardware, or by running a video card in software (slow). Contrast to a Linux VM, which ends up using a pseudo terminal.
The main reason I consider this to be moronic, though, is how completely unnecessary it should be. It speaks volumes about the brittleness of the software. Even if it had no practical implication -- and I believe it does -- what's a good reason for requiring a video card to be present? For not being able to simply run an RDP server, or cmd.exe over a serial console?
Well, because it might be resurrected. We're still looking for funding, though somewhat less vigorously. The layoffs are "temporary".
If we do get funding, it's possible there won't be enough money to hire everyone back. Again, I could point fingers, but I won't.
And I'm not saying everyone suddenly got nice when we went down -- we were like this most of the time. When we'd be late, when we'd push deadlines back, no one was singled out, certainly not as a CYA measure. There simply wasn't time to play political games when we have that much work to do.
many people don't have a luxury of quitting - they have loans, mortgage, family and so on.
Then play the political game, and get ahead. Maybe that will help you do your actual job.
There's a term for that -- "Wage-slave." I hope to never be one, and that's exactly why.
And whether or not you've 'gotten beyond what other proprietary offerings can get you', you've still entered the world where MySQL itself is not $0 cost.
My point here is that it's incredibly unlikely that you've gotten to that point. As far as I know, most proprietary databases will never so much as disclose source code, let alone make it possible to distribute modified binaries.
In order to fall into the case where you need a proprietary MySQL license, you must:
- Be distributing MySQL
- Have modified the source to either the server, or some of the tools
- Refuse to disclose the source to those modifications.
Alternatively, you could:
- Be distributing MySQL
- Want to hide the fact that it's MySQL, so you don't provide any reference to the source code
The absurdly vast majority never even get to the point where they need to redistribute MySQL outside their organization, which means the GPL never kicks in -- you aren't even required to accept it, in that case.
let's remember to count the cost of these other components of open source, to make the comparisons to closed source products more fair and complete,
Well, MySQL is GPLv2. The Linux kernel is also GPLv2. It's kind of a popular license -- I see no reason to warn about MySQL specifically, unless you're going to argue that people more frequently need to modify it.
and not get tricked by someone who misunderstands things like the support requirements of Windows.
So, I'm going to argue that free support via the mailing list and community is actually better than what your $50 Windows license gets you. If you need more than that, it's going to cost you either way.
I suppose $0/box is misleading, but not in the context in which it was presented. $5k just for licensing seems a bit extreme when the alternative is $0/box, just for licensing.
Yes, there are other costs involved, but we know that -- it costs money to buy the components, labor to build them, electricity and space to run them, and time to maintain them -- but it's kind of nice to start thinking about those other costs with an extra $5k to spend.
They also cost bandwidth on Windows. And if you point to WSUS, I'll point to apt-proxy and friends.
commercial support costs license money
There you go. But is any commercial support included in the $50/box cost of Windows?
some Linux compatible software is licensed in ways requiring payment for commercial use. (The MySQL licenses and their interesting clauses come to mind.)
WTF? MySQL can be had under the GPL. By the time you'd need to use the commercial license, I think you're far beyond what other proprietary offerings will get you.
A workable replacement for Outlook, Exchange, and its calendar service is released.
What do you consider "unworkable" about the alternatives? Off the top of my head, there's Gmail + Google Calendar. There's also at least two open source alternatives that I can think of -- either a full stack, or with Outlook as a client.
Swapping back and forth between OpenOffice and Word still causes nasty layout and compatibility problems
From what I've seen, these are exaggerated. Yes, there are problems, but they don't affect most cases. For each worker, there's the question of whether they would actually gain anything from Office, and if so, whether it is worth spending hundreds of dollars on a personal copy for them.
linux professionals are harder to come by and cost more
You get what you pay for. Good Windows admins are harder to come by, and cost more. And a good Linux admin can do more -- manage more machines, spend less time doing it.
they also represent a large risk of taking secret knowledge with them.
And this is different than Windows admins, how?
to be fair to MS, the reason business chooses them is they are cost effective, not because they are the cheapest.
Almost. Business choose them because they believe them to be cost-effective. It's difficult to have an unbiased study back up either as more cost-effective.
does anyone seriously believe windows 2003 with sql server 2005 is a bad platform? i'd suggest if you do you've never used it.
I don't have to use it to think that requiring a video card on a server is fucking moronic. And there are plenty of other reasons to dislike it -- the most recent of which is the 10% premium on services like Amazon EC2.
Slashdot Mirror
You can right-click where it doesn't have any programs or on the edge, and there's a rectangle you can click+hold and drag to change size I think.
Let me put it this way: Once I unlock the widgets, I can open a "configuration" system. In there are about five or ten controls with absolutely no means of labeling them -- not even tooltips.
Ah, I see it. So I made my panel smaller, and almost everything works... Except the clock is broken. It was always broken, but at least I could read the time before. Now the time is chopped off on the top, and the date is chopped off left/right.
WTF? Did anyone even test this?
2.0, to be blunt, stunk, and it really turned me off to KDE4 since 1.4 won't start due to different audio architectures.
See, I have 1.4 working just fine, except for the bug I mentioned. The WONTFIX because you WILL upgrade to Amarok2, which is missing half the features.
And there are a lot of things I really like that were done really well.
I'll agree there. Personally, the most obvious thing I'd miss is alt+f2 (it was too easy to make katapult screw up); okular (kpdf scrolls by some amount less than a page, can't rotate the display, etc); and Konqueror4 just works with a lot more pages (Pandora, for example).
In fact, that alone is a big deal -- that plus Flash 10 means that, for example, I can just fire up a Konqueror window on Pandora and leave it running. On Hardy, I'd have to use Firefox, which means if Flash screws up in some other Firefox window, or if Firefox crashes, the music goes away.
Not sure what you mean about saving a file or browsing porn... though one other feature I kind of miss is the ability to click+drag a link to a konsole, and have it paste a kfmclient command, not just the URL. That way, it goes into KDE's cache, and gives me a nice GUI progress bar. As it is, I have to choose between pasting it there and typing 'wget', or navigating back to that folder via the GUI -- lame.
as of 4.2beta2 they've got about 90+% of "it" nailed as well as or better than 3.5.
Unfortunately, the other 10% are things people really need working.
They've somehow done the 80/20 rule, backwards.
I truly think that most of these shenanigans could have been avoided if they'd tested the final RC on 100 people who'd never used the alphas or betas
I think a single person might have made a difference.
The problem is, this is a desktop environment. Sure, I could've stayed, I could've assumed it was a bug in documentation or UI if I couldn't figure something out (and not just my own stupidity). Or I could've spent my time actually using my computer, in a mostly-working fashion, for things like working, you know, for money.
But the real question here is: Did no one on the Amarok core team actually use it for long enough to figure out how to make a fucking pause button work? Did no one ever actually use the clock applet in the panel?
That's the thing, in Intrepid you can't chose KDE 3.5
I could grab a PPA, or I could grab Hardy. Or I could use Intrepid and KDE4, but fall back to KDE3 for half the packages -- Amarok2 isn't done, so Amarok will be using the KDE3 libs, as will networkmanager (thus breaking kwallet integration)...
it isn't that far behind and it has some added interesting features.
And yet, the fact that it is behind is a big slap in the face. I've told people not to upgrade to KDE4, and I'm about to start telling them to go back to Windows. It's that bad.
I mean, really. Git's good -- git's even really good -- but what does it matter if Perl 6.x is gonna take longer than Duke Nukem Forever to come out?
Maybe Git will help it come out sooner. I know I'd much sooner contribute if I could just fork the project on Github, as opposed to sending patches against a Perforce repository.
And be careful with the DNF jokes. We said the same about Vista, and look what happened.
I'm not a big fan of ESR, but I have to admit "release early and release often" is something I happen to agree with.
Then you obviously missed that point. "Release early and often" is about making the code available, and there are beta/prereleases available for Perl6, so they've got that covered.
It absolutely should not apply to final versions. Those should be released when it's done, not before. For a good example of what happens when people release something just to release it, look at KDE4.
I would guess it's ubiquity and featureset.
Git is built of a patchwork of C and scripts, meaning it's something Perl6 could be a part of someday, and it's also something that's going to be quite familiar to all Perl developers, not just the Pugs guys.
And, Git seems to be quickly becoming the Subversion of DVCS -- fast, open source, everyone has it, everyone knows it, and the alternatives really don't have much compelling to offer.
There are significant advantages of Git over Subversion. RTFS for some.
Just to add insult to injury -- often, a Git checkout, which includes all history, takes up less space than a Subversion checkout for the same project, which doesn't even include recent commit log messages.
But think about this -- you're saying they should use a big, slow, central server, as a single point of failure, crippling offline development, complicating branches (especially merges), and several orders of magnitude slower for just about every operation, just so you don't have to learn a "weird" tool?
See, and I thought it might have something to do with separating work and personal correspondence. Or maybe with branding -- every time you email someone on behalf of the company, you're both advertising the company, and providing some amount of assurance that you work for them.
Or maybe to manage mailing lists, or group calendars. Or to provide email and webmail to people who don't already have them.
The main thing is you can take regular backups of your exchange server. It is not so easy to backup GMail.
Aside from the fact that Google presumably keeps backups, and you're generally not deleting email in the first place, I don't see how it's easier to back up exchange server than Gmail.
It's just IMAP.
I should qualify it, of course -- "different" sometimes is quite a lot better, and more intuitive, even revolutionary. I love KDE4's new alt+f2.
However, a lot of this just feels different for the sake of being different. I don't even mind the lack of files on the desktop -- best example is the panel configuration.
Not sure about GP, but I have. Here's a rant.
Disclaimer: I am using KDE4. I like it for what it could be. As it is, I'm looking at alternatives.
Replace "4.0" with "Vista", "4.1" with "Vista SP1", and "4.2" with "Vista SP2" -- and, for good measure, "3.5" with "XP Pro", and you have a fair sense of what's going on here.
In fact, Microsoft has handled this better -- they still fix bugs in XP.
In KDE4, and in some of the bigger KDE4 apps (like AmaroK), there's this completely new, exciting, amazing version which almost has all the features you needed from the old version, in a very cool-looking but annoyingly different way, and sometimes crashes. Then there's the old, boring, unsupported version, which does everything you want it to do, but has some annoying bugs and deficiencies -- yet whenever you point them out, people close the bug "wontfix" as development has stopped on that branch, and the KDE4 version will be done so differently the bug is irrelevant.
At least Windows has a mostly-working version -- XP. KDE has no working version.
An example of something that worked in 3, but is broken in 4: The panel. Everyone always said, "Don't mind that, it's fixed in 4.1." Well, I'm running 4.1, and I can tell you, it's not even close. How do I make the panel thinner vertically? How do I adjust its translucency -- how do I give it a completely transparent background, but solid foreground?
An example of something that doesn't work anywhere (wontfix in 3, not done yet in 4) is encoding scripts in AmaroK. There's no longer a GUI option to tell AmaroK what your preferred format for a device is -- if you've got an iPod, it's going to give you mp3s, whether you want them or not, even if you can handle AAC just fine. Yet the KDE4 version of AmaroK doesn't yet support encoding scripts in any way, so my choice is mp3s, or no encoding at all. WTF?
Maybe I'm just using the wrong distro? I was pretty appalled at Kubuntu's handling of Intrepid. Bluetooth is broken, due to conflicting versions of a few packages. The only available solutions are, use the commandline (I tried, didn't work), go back to Hardy, or use the Gnome bluetooth GUI.
Isn't that why you use a distro in the first place? So bullshit like this doesn't happen?
Here's hoping by 4.5, they'll finally attain the functionality of 3.5. Maybe they'll still have some users left by then. Meanwhile, I'm going to take a long, hard look at going back to Fluxbox or straight Compiz.
Consider this: every time the Rails team changes things, you'll have to go through your patches and make sure they still apply correctly. And if, heavens forbid, they do some major refactoring, you'll have to spend the time figuring out what functionality got moved where, and re-apply the patches as necessary.
So write plugins, instead of monkeypatching.
Better yet, use a framework like Merb, which has a well-specified Plugin API. Stick with that, and your plugin will work for at least the rest of Merb 1.x. If there's something you can't do with only the plugin API, that is considered a bug.
Merb 2.0 will also be Rails 3.0, so alternatively, wait until then -- Rails will then have a plugin API also.
Couple of points:
Of course, caching and setting up distributed workload processing queues will be necessary in any framework, but that's up to the development team working on a specific app rather than the framework itself.
True -- but Rails does make caching easy.
You may want to look at how Rails 3.0 (the next release of the Rails framework) is incorporating several ideas from Merb to make the framework lighter-weight and more extensible
It's more than that -- Rails 3.0 is also Merb 2.0. So, Rails 3.0 should gain all of the advantages of Merb -- one of which is raw performance.
For offline processing in a Rails app, look at plugins like workling/starling and other message queues.
For what it's worth, Merb has a simple threaded backgrounder built in.
The KDE4 guys too, please...
Also, kindly remind them that "different" isn't always "better", or "more intuitive", or "revolutionary". Sometimes it's just annoyingly unfamiliar, for no good reason.
A mail client that uses an open source calendar source, and integrates it well with email, would be a great Outlook replacement, but I've not seen this either.
Have you tried Kontact?
Last I checked, Kmail couldn't quite handle folders as large as Thunderbird. Other than that, there's Thunderbird with Sunbird.
Is Gmail and Google Calendar working well?
Generally, yes.
Even if it is, it raises the problem that your calendar and email are off-site and you are vulnerable to data theft and loss of services if your external connection is cut.
Haven't had the data theft happen. And unless you're talking about Google stealing data themselves, it seems no more likely than the same thing happening with your own groupware solution, assuming you allow remote access.
I suppose, technically, I could simply connect via IMAP (with a large cache, and offline browsing enabled) and ical+dav. In practice, I've been using Google Apps at work for over a year, and I've never had it go down -- and the times our connection has died, I can count on one hand. But then, it's also less of a concern -- I know I don't get much work done without Internet.
it's still a web client and still slower to work then Outlook.
Citation needed. I've had problems with it, but slowness was never one of them.
Second, many companies are not willing to turn over their email to outside party that they cannot control what they do with it.
My assumption was, there are very specific things they're allowed to do with it.
Besides which, the alternative is trusting Microsoft to not patch Outlook to send them copies of your inbox. If you don't suspect Microsoft of doing that, why do you suspect Google of doing something similar?
Zimbra is a nightmare
That's one of four that I was able to find. Haven't used any, though.
there is no reason to use Outlook as client but not use Exchange as backend.
The reason is, you want it to be a seamless upgrade. For example, replacing the Windows fileserver with Samba on Linux, done right, no one should notice except the admins. Similarly, adding webmail to an IMAP server, even if you have to change the underlying server quite a lot, is something the users shouldn't notice, unless they happen to like webmail.
So, if you have a legitimate reason to want to use something other than Exchange as a backend -- maybe better Linux compatibility, maybe it's easier to admin, maybe it has features Exchange doesn't -- it would be nice if the people using Outlook don't have to care. That's why you want to support Outlook as a client.
and why exactly would a linux admin of equal skill get more done? greater l33tness factor or something?
This is always the counterargument, but it is consistently true. I'm guessing it's got to do with Linux being easier to admin, particularly easier to script, but I don't know a lot about Windows admin.
windows typically utilises standard gui management tools. linux admin's typically utilise cryptic scripts that take even experienced admin's an hours to decypher.
That might be part of the above point on efficiency, then. If you're not scripting repetitive admin tasks, you're Doing It Wrong. And by doing it right, you free up some time to actually document what you do -- which is important no matter how many GUI tools you use.
Now, if you are scripting things, there's not really a difference here -- on Linux, you've got standard commandline management tools; on Windows, they're standard GUI management tools. Do you script that GUI? If so, I'll bet your scripts are even more cryptic and fragile than the worst Linux admin script. If not, you're wasting a lot of time on repetitive bullshit you shouldn't have to do.
well, i can pick up the phone and have MS tell me exactly what my volume license costs me and what i get with it. can you do the same with linux?
No, MS doesn't do Linux.
But seriously, yes, I can call Canonical, or I can call Redhat, or even Novell. Or, I can do the support in-house, if needed. Or I can hire a contractor and get vague estimates.
Point is, there's a choice. If you're doing Windows, you pick up the phone and call MS, full stop. If there's something the OS doesn't do that you want it to do (even a simple security hole), you call MS and grovel, or you pay through the nose for "Shared Source" and someone to code it, assuming the parts you want are even available that way.
why is it moronic? if you lose network connectivity what do you intend on doing? serial terminal?
Why not? Plug it into a neighboring machine, and hook that up to the network. Much faster than doing the same with a capture card of some sort.
If it's unbootable, pull the hard drive, put it in another machine.
you may as well have a video card since you can't buy a mobo without one these days.
Doesn't have to be bare metal -- what about a virtual machine?
Supposing it is bare metal, you apparently weren't looking. Why should I have to buy a video card on top of that?
Why is it any different to "requiring" a serial port or a Lights-out-management card ?
Mostly because a serial terminal can be managed remotely, and because that video card takes up more space and electricity than a serial port.
And yet, I seem to remember not needing a serial port either. It's a headless machine; let it be headless.
Incidentally, I doubt the vast majority of hardware engineers are "moronic", yet for some reason they think including a video card is a reasonably good idea.
Must be a recent phenomenon, then. I certainly remember buying motherboards without video cards, which would run fine without them. If it was unbootable for some reason, you'd go find a spare PCI video card and use that.
And yet, we're not necessarily talking about hardware, either. Consider virtual machines, and the problem compounds. You now need a virtual display somewhere, either by virtualizing access to the actual video hardware, or by running a video card in software (slow). Contrast to a Linux VM, which ends up using a pseudo terminal.
The main reason I consider this to be moronic, though, is how completely unnecessary it should be. It speaks volumes about the brittleness of the software. Even if it had no practical implication -- and I believe it does -- what's a good reason for requiring a video card to be present? For not being able to simply run an RDP server, or cmd.exe over a serial console?
Actually, I pay the bills with the money I saved from the job I had, which was a job I wanted to have.
And, at the end of the day, I'd much rather eat ramen than be a wage slave.
One could point fingers, but why?
Well, because it might be resurrected. We're still looking for funding, though somewhat less vigorously. The layoffs are "temporary".
If we do get funding, it's possible there won't be enough money to hire everyone back. Again, I could point fingers, but I won't.
And I'm not saying everyone suddenly got nice when we went down -- we were like this most of the time. When we'd be late, when we'd push deadlines back, no one was singled out, certainly not as a CYA measure. There simply wasn't time to play political games when we have that much work to do.
many people don't have a luxury of quitting - they have loans, mortgage, family and so on.
Then play the political game, and get ahead. Maybe that will help you do your actual job.
There's a term for that -- "Wage-slave." I hope to never be one, and that's exactly why.
And whether or not you've 'gotten beyond what other proprietary offerings can get you', you've still entered the world where MySQL itself is not $0 cost.
My point here is that it's incredibly unlikely that you've gotten to that point. As far as I know, most proprietary databases will never so much as disclose source code, let alone make it possible to distribute modified binaries.
In order to fall into the case where you need a proprietary MySQL license, you must:
- Be distributing MySQL
- Have modified the source to either the server, or some of the tools
- Refuse to disclose the source to those modifications.
Alternatively, you could:
- Be distributing MySQL
- Want to hide the fact that it's MySQL, so you don't provide any reference to the source code
The absurdly vast majority never even get to the point where they need to redistribute MySQL outside their organization, which means the GPL never kicks in -- you aren't even required to accept it, in that case.
let's remember to count the cost of these other components of open source, to make the comparisons to closed source products more fair and complete,
Well, MySQL is GPLv2. The Linux kernel is also GPLv2. It's kind of a popular license -- I see no reason to warn about MySQL specifically, unless you're going to argue that people more frequently need to modify it.
and not get tricked by someone who misunderstands things like the support requirements of Windows.
So, I'm going to argue that free support via the mailing list and community is actually better than what your $50 Windows license gets you. If you need more than that, it's going to cost you either way.
I suppose $0/box is misleading, but not in the context in which it was presented. $5k just for licensing seems a bit extreme when the alternative is $0/box, just for licensing.
Yes, there are other costs involved, but we know that -- it costs money to buy the components, labor to build them, electricity and space to run them, and time to maintain them -- but it's kind of nice to start thinking about those other costs with an extra $5k to spend.
Updates cost bandwidth
They also cost bandwidth on Windows. And if you point to WSUS, I'll point to apt-proxy and friends.
commercial support costs license money
There you go. But is any commercial support included in the $50/box cost of Windows?
some Linux compatible software is licensed in ways requiring payment for commercial use. (The MySQL licenses and their interesting clauses come to mind.)
WTF? MySQL can be had under the GPL. By the time you'd need to use the commercial license, I think you're far beyond what other proprietary offerings will get you.
When bad things happen people search for someone to blame.
I say again: The problem is not that you failed to cover your ass. The problem is that you're in a job where you feel the need to.
I was in a startup that recently imploded. It would have been easy to start pointing fingers, but no one did.
A workable replacement for Outlook, Exchange, and its calendar service is released.
What do you consider "unworkable" about the alternatives? Off the top of my head, there's Gmail + Google Calendar. There's also at least two open source alternatives that I can think of -- either a full stack, or with Outlook as a client.
Swapping back and forth between OpenOffice and Word still causes nasty layout and compatibility problems
From what I've seen, these are exaggerated. Yes, there are problems, but they don't affect most cases. For each worker, there's the question of whether they would actually gain anything from Office, and if so, whether it is worth spending hundreds of dollars on a personal copy for them.
Now your choice is either the slavery office-space style, or go to the free soup line at salvation army...
Or unemployment benefits, and then a choice of companies.
linux professionals are harder to come by and cost more
You get what you pay for. Good Windows admins are harder to come by, and cost more. And a good Linux admin can do more -- manage more machines, spend less time doing it.
they also represent a large risk of taking secret knowledge with them.
And this is different than Windows admins, how?
to be fair to MS, the reason business chooses them is they are cost effective, not because they are the cheapest.
Almost. Business choose them because they believe them to be cost-effective. It's difficult to have an unbiased study back up either as more cost-effective.
does anyone seriously believe windows 2003 with sql server 2005 is a bad platform? i'd suggest if you do you've never used it.
I don't have to use it to think that requiring a video card on a server is fucking moronic. And there are plenty of other reasons to dislike it -- the most recent of which is the 10% premium on services like Amazon EC2.