Indeed, something similar happened when at I was Sun. With modern servers you just have to have an IPMI client for remote lights-out management. The most popular one is IPMItool, an OS product that got support on Unix-like systems early on. But somebody managing a remote system might well be running Windows. IPMItool will run on Cygwin, but Sun can't redistribute Cygwin, so they needed to provide customers with a native Windows version. For that, they hired a software consulting firm to make the port, then released the source code in accordance with the original software license.
The difference here is that Sun is a very bureaucratic place where you can't do anything without jumping through all the right hoops. So if you use OS software or code, you're required to tell the company lawyers so they can make sure you don't break any rules. At other places I've worked, it was pretty common for some engineer to see some OS code he wanted to borrow and just go ahead and use it. Any OS license requirements might be ignored or the engineer might try to interpret them on his own, with the resulting mistakes that amateur lawyers always make. Either way you have a violation of the OS license that has more to do with stupidity than with any grand conspiracy. A classic example of Hanlon's Razor.
You do not want a consumer device with an optical drive built in, unless the drive is the device. It's always the first thing to fail. That's not a big deal on computers, where such drives are commodity OEM products, so replacing them is not an expensive send-it-back-to-the-factory process. But everybody I know who's bought a TV, VCR, or game console with a blu-ray built in has regretted it.
Since this box looks extremely hackable, you could probably add a blu-ray drive to it. Problem solved.
'm waiting for someone to get XBMC going on a PS3.
Why does it have to be a PS3? Why not just run XBMC on a Linux box with a blu-ray drive?
(Note for those backwards people who still use POP: IMAP users normally download their message bodies on the fly.)
Sigh. Been using the beta for a couple of weeks, so I'm familiar with this download-everything behavior. This is not actually a new feature. What's changed is that it's enabled by default. Which is, I agree, pretty dumb.
Here's why they did this. This version has vastly improved searching (far and away, my favorite new feature) which doesn't work unless you have a local copy of the mailbox for indexing.
(I find this a godsend. In the past, I've turned on the local copy feature and then used Google desktop search. The problem here is that the user interface for GD sucks. Also, on one of my machines, I can't get GD to even look at the local mailbox file — no idea why.)
The way Firefox 3 does searching is Ultimately Kewl. (Won't try to describe it, go give it a try.) Naturally, they were proud of this feature and wanted everybody to try it. But just enabling such a potential bandwidth raper was dumb. Somebody should have designed a wizard or something so you could select the mailbox folders you wanted to index.
I am so tired of hearing "e-readers will never replace books" arguments, as if it were an all-or-nothing thing. I can well imagine stone carvers makers the same "permanence" argument against books.
E-readers still can't do a lot of what books do, but so what? Half the books I read, I read once, then give them away or return them to the library. For these, an e-reader is perfectly fine. And as the technology advances, a physical book will have fewer and fewer advantages.
Frankly, I think all this strident ranting against e-books is just people resisting having to learn new ways of doing things. Which is fine for them, but why must they lecture the rest of us all the time?
And as a writer myself, I have very little patience when this attitude shows up in the people I work with. In particular, it's a pain when editors and reviewers insist on physical copies so they can scribble comments in the margin. So then I have to decipher their handwriting and cryptic comments. And once, when I did an actual mass-market book, the publisher's editor and I had to FedEx pages back and forth, at great cost in time and money. Learn to use Acrobat, people!
Excuse me, where did I say that you should call 911 if you're being bullied? The guy I responded to made a lame comparison between asking for help and calling 911. That's what I was flaming.
And don't strawman my argument. I never said that there were no resources for people who need help. I said that people who belittle the idea of asking for help prevent people from doing so.
Your attitude is exactly why people don't ask for help when they're in a bad place. The result is school shootings, suicides, and other depressing events. Not just among teens either. Adults in general listen to your kind of belittlement and think they're weak if they ask for help.
This is a particular problem in first responders and members of the armed services. They see all kind of horrible crap, need help in dealing with it, and are afraid to get it because they don't want to seem "weak". This is a particular problem in the Army, which is seeing a spike in suicides lately.
Mind you, I'm not equating a bullied teenager with a GI who's seen his friends blown up by an IED. But they do have one thing in common: they need for it to be easy to ask for help, and people like you make it hard.
I still don't see how you can reduce the deployment of LogMeIn Free to sending your clients a download link. I can see where Central simplifies things at your end. But your client still has to login to your LogMeIn account, download the client, and reboot. What am I missing?
Its free for business use as well. Their eula does not forbid it, suprisingly enough.
LogMeIn Free is come-on for their other remote desktop products. In a sense it's crippleware, since everything beyond basic remote control is disabled.
Which is not a criticism. Actually, it's pretty cool that they allow people to use their servers for free. But some of the features they leave out are precisely the ones I'd like to have to support friends and family that call me up for help. (Like one-click deployment on the computer you want remote access to.) To get that feature, you have to pay $129/month for LogMeIn Rescue. That's a reasonable price if you're running a help desk, but too expensive if you just use it occasionally.
Deployment link? Have you actually used this software? To use the free version of logmein, the person you're trying to help has to log in using your password, then download and configure the client to work in your account. It's doable, but it requires some non-trivial handholding.
Perhaps you're thinking of LogMeIn Rescue. I've never used it, but it's probably as simple as they claim it is. (This company appears to have pretty high standards.) One little problem: it's not cheap.
I use Logmein for this purpose too (and agree that it's really good software), but it's a little early to close the discussion. Installing the free version on every machine you want to support is a pain, especially if you have to talk somebody else through it over the phone. There's a simpler "rescue" version, but it's not free, or even cheap.
The latest version Skype has a desktop sharing feature. No remote control, but if you can see the other screen it's easy to talk people through most tasks. Still a pain to install, but at least Skype is something people have a day-to-day use for.
Like I said, I'm sure license fees are a factor (especially when you're a small independent developer). I'm only saying that it's not a dominating factor.
Did it? Embedded Linux certainly does well, but I don't see it dominating the marketplace. How many times have you passed a Jumbotron and seen a Windows error message instead of an ad? The lack of license fees for Linux is certainly a factor in the embedded market, but I don't see it as a deciding factor.
I used to work at Sun, documenting servers. Nowadays, all Sun servers have little Service Processors for remote and "lights-out" management. The current generation of these are all ARM or PPC Linux systems. Linux is ideal for this kind of application: small, elegant, and it's easy to remove all the crap you don't need. But suppose Linux had not been ideal for this particular app? Suppose that some non-free OS could have been adapted with less development work on Sun's part. (Probably not Windows, but there are lots of embedded OSs out there.) It would make no sense then to use Linux just to save a $10 license fee. Even having to hire even a single extra programmer would eat up all the savings.
I have a Windows Mobile phone. My previous phone was Palm Garnet. Neither of these presumed to tell me which apps I could install. Android is probably a more advanced OS than either of these, but I don't see how anybody's managed to leverage that into a superior user experience. Hence my question.
You're the third response to my post that has the unstated assumption that Android is the only real alternative to iPhone. Not true: besides the two I mentioned, there's Symbian, Blackberry, Linux, and some others. Most of them have been around a lot longer than iPhone. It seems that even people who don't buy Apple products are sucked in by their mystique.
It's free (as in cost), an established standard, and backed by a company that's very likely still going to be around in a few years. These are all reasons to use it if you're producing mobile phones.
That argument was used in favor of desktop Linux too. It's one factor in favor, there are many other factors against.
The market isn't overcrowded. iPhone has something like a 2.5% market share. At least some of those remaining 97.5% are going to be upgrading to a smartphone.
What, you think iPhone OS is the only established smartphone OS? Not true. In fact, it's something of a newcomer.
Consider this factor: lots of people think that blowing up the planet is a really bad idea. No Armageddon, no need for Armageddon weaponry. Such Ludditism may shock you, but there it is.
That's one obvious reason. Another is the huge trendiness of the iPhone. And let's not forget all those "there's an ap for that" commercials.
I have to say, I don't get Android. What's the appeal? Why does anybody think it can make headway in an already-overcrowded mobile OS market? Just because it's Open Source? If so, it's the Linux Desktop Uprising all over again. You know, that period about 8 years ago when there was so much excitement about Linux displacing Windows, completely ignoring Windows' insurmountable lock-in factor.
It's great that unelected bureaucrats in California are clamoring to save energy, but when they target your big-screen TVs for elimination
Why did you submit this article? You obviously didn't read it. There's nothing about banning big screen TVs, just a bit about tightening up energy requirements a tad.
My guess is that you saw an angry, ignorant rant on somebody's blog and are now parroting it. You're an illiterate fool.
OK, your practical experience certainly trumps my theorizing. But I'm going to theorize anyway. If the result is stupid, please go easy on me.
As I understand it, ZFS supports variable block size, which should avoid the kind of small-file wastage you describe. Is it possible that your problems with ZFS has more to do with your inexperience with it than with its technical limitations?
It's worth noting how much easier you found it to set up Samba. I personally find Samba a nightmare to administer — the learning curve is way too steep. Now, if Samba were something I'd had to deal with a lot (as apparently it is for you), I'd probably find it a lot more intuitive, despite its complexities. But if I had to try setting up an SMB server like yours, I'd certainly try Solaris CIFS first on the off chance that the learning curve isn't as nasty.
What I'm getting at is a principle I often see reflected in Slashdot posts: software you're experienced with is subjectively simpler, even if it's objectively far more complex. (Look at the text editor wars for an example.) I don't know the software involved enough to say for sure that this is the case here, but it sure sounds like it.
Your story is very familiar. Almost every day, a big company buys a little company thinking that a successful small business can become a blockbuster big business with a little infusion of capital and other resources. Some companies know how to pull this off, but it usually seems to fail, both because of the scaling issues you describe and because of the clash of management cultures between the two entities.
I used to work at Sun, and that company made one disastrous acquisition after another. The last one was particularly ironic: MySQL. It took Sun almost no time at all to drive away the key MySQL people (something that should have been predicted given their profound difference in corporate cultures) rendering the acquisition almost worthless. Now, Sun wants to be acquired itself by one of the few companies that knows how to make an acquisition pay — only to see the deal prevented or fatally delayed by EU antitrust enforcers who don't want two leading RDBMSs under one roof.
Slashdot Mirror
Indeed, something similar happened when at I was Sun. With modern servers you just have to have an IPMI client for remote lights-out management. The most popular one is IPMItool, an OS product that got support on Unix-like systems early on. But somebody managing a remote system might well be running Windows. IPMItool will run on Cygwin, but Sun can't redistribute Cygwin, so they needed to provide customers with a native Windows version. For that, they hired a software consulting firm to make the port, then released the source code in accordance with the original software license.
The difference here is that Sun is a very bureaucratic place where you can't do anything without jumping through all the right hoops. So if you use OS software or code, you're required to tell the company lawyers so they can make sure you don't break any rules. At other places I've worked, it was pretty common for some engineer to see some OS code he wanted to borrow and just go ahead and use it. Any OS license requirements might be ignored or the engineer might try to interpret them on his own, with the resulting mistakes that amateur lawyers always make. Either way you have a violation of the OS license that has more to do with stupidity than with any grand conspiracy. A classic example of Hanlon's Razor.
ICBMs can be tested.
Provided you leave off the warhead!
You do not want a consumer device with an optical drive built in, unless the drive is the device. It's always the first thing to fail. That's not a big deal on computers, where such drives are commodity OEM products, so replacing them is not an expensive send-it-back-to-the-factory process. But everybody I know who's bought a TV, VCR, or game console with a blu-ray built in has regretted it.
Since this box looks extremely hackable, you could probably add a blu-ray drive to it. Problem solved.
'm waiting for someone to get XBMC going on a PS3.
Why does it have to be a PS3? Why not just run XBMC on a Linux box with a blu-ray drive?
(Note for those backwards people who still use POP: IMAP users normally download their message bodies on the fly.)
Sigh. Been using the beta for a couple of weeks, so I'm familiar with this download-everything behavior. This is not actually a new feature. What's changed is that it's enabled by default. Which is, I agree, pretty dumb.
Here's why they did this. This version has vastly improved searching (far and away, my favorite new feature) which doesn't work unless you have a local copy of the mailbox for indexing.
(I find this a godsend. In the past, I've turned on the local copy feature and then used Google desktop search. The problem here is that the user interface for GD sucks. Also, on one of my machines, I can't get GD to even look at the local mailbox file — no idea why.)
The way Firefox 3 does searching is Ultimately Kewl. (Won't try to describe it, go give it a try.) Naturally, they were proud of this feature and wanted everybody to try it. But just enabling such a potential bandwidth raper was dumb. Somebody should have designed a wizard or something so you could select the mailbox folders you wanted to index.
I am so tired of hearing "e-readers will never replace books" arguments, as if it were an all-or-nothing thing. I can well imagine stone carvers makers the same "permanence" argument against books.
E-readers still can't do a lot of what books do, but so what? Half the books I read, I read once, then give them away or return them to the library. For these, an e-reader is perfectly fine. And as the technology advances, a physical book will have fewer and fewer advantages.
Frankly, I think all this strident ranting against e-books is just people resisting having to learn new ways of doing things. Which is fine for them, but why must they lecture the rest of us all the time?
And as a writer myself, I have very little patience when this attitude shows up in the people I work with. In particular, it's a pain when editors and reviewers insist on physical copies so they can scribble comments in the margin. So then I have to decipher their handwriting and cryptic comments. And once, when I did an actual mass-market book, the publisher's editor and I had to FedEx pages back and forth, at great cost in time and money. Learn to use Acrobat, people!
Right let's kill off the mentally ill before they cause problems. Let's take care of the Jews and the Gypsies while we're at it. Heil Hitler!
Excuse me, where did I say that you should call 911 if you're being bullied? The guy I responded to made a lame comparison between asking for help and calling 911. That's what I was flaming.
And how is clicking on an icon like calling 911?
And don't strawman my argument. I never said that there were no resources for people who need help. I said that people who belittle the idea of asking for help prevent people from doing so.
And there's a big difference between clicking an icon and asking for help and calling 911.
Your attitude is exactly why people don't ask for help when they're in a bad place. The result is school shootings, suicides, and other depressing events. Not just among teens either. Adults in general listen to your kind of belittlement and think they're weak if they ask for help.
This is a particular problem in first responders and members of the armed services. They see all kind of horrible crap, need help in dealing with it, and are afraid to get it because they don't want to seem "weak". This is a particular problem in the Army, which is seeing a spike in suicides lately.
Mind you, I'm not equating a bullied teenager with a GI who's seen his friends blown up by an IED. But they do have one thing in common: they need for it to be easy to ask for help, and people like you make it hard.
I still don't see how you can reduce the deployment of LogMeIn Free to sending your clients a download link. I can see where Central simplifies things at your end. But your client still has to login to your LogMeIn account, download the client, and reboot. What am I missing?
Its free for business use as well. Their eula does not forbid it, suprisingly enough.
LogMeIn Free is come-on for their other remote desktop products. In a sense it's crippleware, since everything beyond basic remote control is disabled.
Which is not a criticism. Actually, it's pretty cool that they allow people to use their servers for free. But some of the features they leave out are precisely the ones I'd like to have to support friends and family that call me up for help. (Like one-click deployment on the computer you want remote access to.) To get that feature, you have to pay $129/month for LogMeIn Rescue. That's a reasonable price if you're running a help desk, but too expensive if you just use it occasionally.
Deployment link? Have you actually used this software? To use the free version of logmein, the person you're trying to help has to log in using your password, then download and configure the client to work in your account. It's doable, but it requires some non-trivial handholding.
Perhaps you're thinking of LogMeIn Rescue. I've never used it, but it's probably as simple as they claim it is. (This company appears to have pretty high standards.) One little problem: it's not cheap.
I use Logmein for this purpose too (and agree that it's really good software), but it's a little early to close the discussion. Installing the free version on every machine you want to support is a pain, especially if you have to talk somebody else through it over the phone. There's a simpler "rescue" version, but it's not free, or even cheap.
The latest version Skype has a desktop sharing feature. No remote control, but if you can see the other screen it's easy to talk people through most tasks. Still a pain to install, but at least Skype is something people have a day-to-day use for.
BTW, from my perspective your own post is more typical of Slashdot than my own:
Ouch!
theorizing without data,
I had some data
postulating ignorance on the part of those with experience,
Hey if I hadn't postulated that you were wrong, you wouldn't have bothered to prove that you're right. Said proof was educational for me.
and generally trying to adopt a tone of undue authority.
Sorry if I came across that way. Not my intent.
Please note that I didn't call you any names. I didn't even mention Hitler! I think that destroys my Slashdot Flame Warrior status.
Like I said, I'm sure license fees are a factor (especially when you're a small independent developer). I'm only saying that it's not a dominating factor.
Dude, allow me some hyperbole. Or didn't you notice my sarcasm?
But it's certainly worked for embedded Linux.
Did it? Embedded Linux certainly does well, but I don't see it dominating the marketplace. How many times have you passed a Jumbotron and seen a Windows error message instead of an ad? The lack of license fees for Linux is certainly a factor in the embedded market, but I don't see it as a deciding factor.
I used to work at Sun, documenting servers. Nowadays, all Sun servers have little Service Processors for remote and "lights-out" management. The current generation of these are all ARM or PPC Linux systems. Linux is ideal for this kind of application: small, elegant, and it's easy to remove all the crap you don't need. But suppose Linux had not been ideal for this particular app? Suppose that some non-free OS could have been adapted with less development work on Sun's part. (Probably not Windows, but there are lots of embedded OSs out there.) It would make no sense then to use Linux just to save a $10 license fee. Even having to hire even a single extra programmer would eat up all the savings.
I have a Windows Mobile phone. My previous phone was Palm Garnet. Neither of these presumed to tell me which apps I could install. Android is probably a more advanced OS than either of these, but I don't see how anybody's managed to leverage that into a superior user experience. Hence my question.
You're the third response to my post that has the unstated assumption that Android is the only real alternative to iPhone. Not true: besides the two I mentioned, there's Symbian, Blackberry, Linux, and some others. Most of them have been around a lot longer than iPhone. It seems that even people who don't buy Apple products are sucked in by their mystique.
It's free (as in cost), an established standard, and backed by a company that's very likely still going to be around in a few years. These are all reasons to use it if you're producing mobile phones.
That argument was used in favor of desktop Linux too. It's one factor in favor, there are many other factors against.
The market isn't overcrowded. iPhone has something like a 2.5% market share. At least some of those remaining 97.5% are going to be upgrading to a smartphone.
What, you think iPhone OS is the only established smartphone OS? Not true. In fact, it's something of a newcomer.
http://en.wikipedia.org/wiki/Mobile_operating_system#Operating_systems
Consider this factor: lots of people think that blowing up the planet is a really bad idea. No Armageddon, no need for Armageddon weaponry. Such Ludditism may shock you, but there it is.
That's one obvious reason. Another is the huge trendiness of the iPhone. And let's not forget all those "there's an ap for that" commercials.
I have to say, I don't get Android. What's the appeal? Why does anybody think it can make headway in an already-overcrowded mobile OS market? Just because it's Open Source? If so, it's the Linux Desktop Uprising all over again. You know, that period about 8 years ago when there was so much excitement about Linux displacing Windows, completely ignoring Windows' insurmountable lock-in factor.
It's great that unelected bureaucrats in California are clamoring to save energy, but when they target your big-screen TVs for elimination
Why did you submit this article? You obviously didn't read it. There's nothing about banning big screen TVs, just a bit about tightening up energy requirements a tad.
My guess is that you saw an angry, ignorant rant on somebody's blog and are now parroting it. You're an illiterate fool.
OK, your practical experience certainly trumps my theorizing. But I'm going to theorize anyway. If the result is stupid, please go easy on me.
As I understand it, ZFS supports variable block size, which should avoid the kind of small-file wastage you describe. Is it possible that your problems with ZFS has more to do with your inexperience with it than with its technical limitations?
It's worth noting how much easier you found it to set up Samba. I personally find Samba a nightmare to administer — the learning curve is way too steep. Now, if Samba were something I'd had to deal with a lot (as apparently it is for you), I'd probably find it a lot more intuitive, despite its complexities. But if I had to try setting up an SMB server like yours, I'd certainly try Solaris CIFS first on the off chance that the learning curve isn't as nasty.
What I'm getting at is a principle I often see reflected in Slashdot posts: software you're experienced with is subjectively simpler, even if it's objectively far more complex. (Look at the text editor wars for an example.) I don't know the software involved enough to say for sure that this is the case here, but it sure sounds like it.
Your story is very familiar. Almost every day, a big company buys a little company thinking that a successful small business can become a blockbuster big business with a little infusion of capital and other resources. Some companies know how to pull this off, but it usually seems to fail, both because of the scaling issues you describe and because of the clash of management cultures between the two entities.
I used to work at Sun, and that company made one disastrous acquisition after another. The last one was particularly ironic: MySQL. It took Sun almost no time at all to drive away the key MySQL people (something that should have been predicted given their profound difference in corporate cultures) rendering the acquisition almost worthless. Now, Sun wants to be acquired itself by one of the few companies that knows how to make an acquisition pay — only to see the deal prevented or fatally delayed by EU antitrust enforcers who don't want two leading RDBMSs under one roof.