One thing that used to annoy me with "getting to my lag-free desktop" (and lets face it, thtas what its all about) was that i'd turn on and have to wait for a login prompt... then login and wait for a responsive desktop. When fed 9 came out I decided to use the luks encrypted disk - this turned out to be quite a massive bonus in some ways because the thing asks me for a password so early in the boot process.
I then turned on auto-login at the login screen and made firefox and evolution startup programs and wah lah my lifes (generally) alot simpler. i.e. turn on computer - 5 seconds later type in luks password - go make tea - come back to responsive desktop with my most used applications already open. A fast post is going to shave 5 seconds off my booting life? I really dont think im going to be that excited.
What I would prefer though is a rock-solid hibernate/restore strategy, and the idea (a little while ago) about a machine that shutdown, boots up and then hibernates does actually appeal but what kills that strategy is my laptop (where it would be useful) cause at home i have no docking station and at work I do and the thing can never hibernate between the two.
The only place i've ever found it useful is on the netbook where im booting up and down a fair bit and the only login is the encrypted disk login (right at the very beginning)... the rest of my machines I really dont care too much about fast boot times.
Actually, how often it boots and fast it boots are often small considerations - depending on where you sit.
Consider a file server that crashes and reboots twice a year and takes 5 minutes to come back up... thats 99.998% availability and from an infrastructure perspective thats pretty dang awesome. From a helpdesk point of view, they'll suffer one heck of a beeting everytime it goes down and that'll be all they remember of the server. Same goes for the users, all they'll remember is the 5 minutes it went down when half the company was doing something important.
Consider active directory though (or any kind of multi-master replicated service) - again, your talking about a server thats really not doing anything terribly difficult, but unlike the file server if it goes down no ones likely to notice. On top of that your often more concerned that the server will come up working then how quickly.
As for noise, well the good ol e450 from sun fully stacked with 20 15k disks (not to mention its in-built fans) used to make very amusing noises as it came up and some of the time it went down was because the disks managed to disconnect from the backplane (and you could hear the difference if you were around it during a boot often enough) - a good hard whack and it was sorted. Wondefull little machines those
Personally, on the server side, i remember the top end of sun's line back in its peek (from the e250, e450 all the way of to the e10k, the e25k, and the sunfire range)
5 minutes would have been a blessing on those machines.
Partly untrue, while the kind of benchmarks in the article have nothing (directly) to do with "Feel" alot of companies who have web sites (especially the big ones) do have tools to gauge feel.
The two i stumble across most often are from Segue and Mercury (Silk perform and load runner and their associated software bits). Seque's silk line is very good at judging feel, its designed to hit websites (as well as DB's and almost anything really) pretty hard with "real user access" patterns and can tell you how it'll perform from the user perspective depending on the browser they're using (it emulates various browsers). It can tell you things like the time taken to first start rendering the page, how long it took to render the page entirely and how long it took to download it all (as well as the impact of various link speeds). Those are the only two im particularly familiar with, but there are many many more. Ultimately, you can take all that data, and get a feel for "feel" based on browser type and link speed.
Coming form an SI type role, seeing what automated testing tools exist these days is pretty kewl (more used by testers then myself) and gives you a good feel of "when im building infrastructure what am I aiming to do".
I agree with your points for the most part... however, android has more in common with the PC then any other platform... plus, back in those days there were no FOSS alternatives (not real ones anyways)...
The reason i say that is similar to how people got into making PC - an open standard. Alot of phone makers would love one that allows them to make a phone that is fairly industry-wide compatible and the old real option at this point is windows.
Everyone would love to be able to build an iphone compatible but apple would never do that (Which is a pity imho cause that would catch on - you could see rim one day making that switch perhaps).
The big plus with android though is its potential for slimming down. Take the biggest owner in the market, nokia, the big sellers in their phones run on small chips that are just good enough to run symbian. Symbians greatest asset is that it can run on a much smaller footprint than windows can and ultimately makes for cheaper hardware. Android has a similar potential.
My point was simply this, android brings alot of potential to the industry, its not there yet, but it will be. consider:
1) its a sexy interface 2) its not exactly slow 3) its highly modifiable 4) it can be cut down to almost nothing 5) anyone can use it 6) its very powerful (most of that I learnt by hacking my tytn2 to run android - horrible phone that tytn2, yet most people say its one of the best windows phones and probably sadly true)
which gives it all the advantages of all the other OS's and some more. 1) sexy like an iphone 2) slim like symbian 3) anyone can build on it like windows 4) its open like... err... openmoko? moblin? 5) its powerfull like iphone and windows mobile. 6) much like j2me phones, you dont have to target the cpu, but unlike j2me it has alot more power.
Anyways, thats my 0.02, its probably mostly wrong.
if im not mistaken, Wind River had two os's... VxWorks (Their original OS and had features like RTOS) and they also had something linuxy that they also had a vxworks brand on it..
I could be wrong though.
VxWorks was certianly nothing like linux though, so how you combine something destined for vxworks and linux together is beyond me... perhaps its jsut a badly worded statement or theres a back story here thats gone missing... maybe they're going to OSS VxWorks and roll up some of its components to linux? Or maybe they're going to make the two interoperable somehow? who knows?
And reading the actual article I can entirely agree with him, but I also think its partly his own fault.
Linus himself (not that long ago) was talking about how they "deliberately" make the kernel code difficult to understand and he had good reasons for that which revolved around not getting people interested unless they're willing to stick with it... i.e. to avoid orphaned code. From my own experience I started writing a kernel module in the scsi stack that i later abandoned because another piece of work expanded to provide the exact same functionality.
The problem though, for me, was that the entire time I was writing my little module it kept occuring to me that there were other parts of the stack I could easily hand off to rather then re-writing it myself. Unfortunately the code is often so hard to read that you spend more time trying to understand if some other segment of kernel code is actually doing what you need to do. Thats painful. Ultimately you end up writing your own little bit of code because you need to be intimately familiar with other branches of code to understand whether you need to implement something or not.
Hence the bloat, im sure there is tonnes of code that could be moved into higher levels of functionality in the kernel that instead get written over and over again even though they may be able to be reused. An example is SCSI and Network, now they are very different branches of code dealing with totally disparate systems but ultimately they do similar things, i.e. take a chunk of data from one interface and dump into another interface. I'd be curious to know how much of what they do gets passed into some higher level kernel code though...?
Raid may be coming to end, but its nothing to do with the speeds of drives. Its all about size. Its all about the fact that you are GUARANTEED to have a sector fail on read after x number of reads. Now because x hasnt changed much over the decades while drives have gotten vastly larger, we're approaching x but we've still got a while to go yet and in the enterprise (where raid actually matters) its not something that'll really be a problem for at least another 5-10 years. This is not a problem that raid 6 or zfs solve.
But, the problem is this, if a drive fails in a raid 5 array and the array runs off to rebuild it then when you get near x you'll fail during the rebuild and that will keep occuring until the array gives up. Raid 6 still has the same issue, it'll still read all the drives to rebuild the broken disk, same with ZFS. Unfortunately, there has been alot of industry spin suggesting raid 6 and ZFS solve this issue which is patently false.
Raid 6 was designed because of how arrays were being deployed, when large scale disk boxes were being deployed (SAN etc) we got to a point where if you had enough disks you might as well have an engineer on site because the failure rate of drives was such that when you got to a certain number of drives (we're talking hundred of thousands here, not that hard in a large data center packed with SAN arrays) you could guarentee at least one drive would fail EVERY SINGLE DAY. So, with raid sets getting larger (i.e. 10-15 drives in a single raid 5 array and they were getting much bigger than that) it became important to save yourself from a dual-drive failure (because the likelyhood of it happening went up exponentially as the set got larger), having two parity drives solved that issue - hence raid 6.
ZFS was designed for performance. Raid 5 (and 6) SUCKS for performance when your not using some form of off-load device (hardware raid card or SAN array head) and so ZFS was designed to (in part) give people the ability to do that kind of thing on the server cpu without killing the performance of having 6-10 striped drives. The problem is that if you throw some data at a drive, computers are very good at it. They can just say "write this chunk of data here please" and the rest just happens. With raid (if its done in software), what you have to do is calculate the parity of ever single byte of data on the array - this hurts alot and destroys the whole transfer mechanism.
The point is though, we're approaching x and the only real solution is to fix x. Though, with SSD's coming along quite nicely, the problem will likely resolve itself and we'll be using raid 6 SSD's. SSD's are quite capable of solving this little drama because x can be quite variable unlike their "spinning-rusty-metal" counterparts. They can also be scaled alot easier for physical size (something we havent seen yet). Its very hard to produce a tonne of different sizes when it comes to disks - i.e. we have three formats, 1.8", 2.5" and 3.25" and alot of that has to do with producing some "spinning-rusty-metal" in an efficient way at different sizes. i.e. thats just not practical for motorized components. SSD's on the other hand are a completely different boat. Because they are just chips on a little board they can scale very well in terms of physical size. Right now it hasnt happened, they are doomed to follow the pre-set sizes of their cousins because of the fact we're forced to use both.
This wont always be the case, eventually not only will SATA, FC, infiniband (an already mostly dead tech any way) and SAS be a thing of the past (They were designed for spinning rusty metal and with people like seagate threatening to sue ssd makers for using sata and the like, ssd boys will probably make their own interface). The traditional interfaces for HD's are not fantastic (as it turns out) for ssd's, you could in fact make a much better interface if your talking directly to ssd's. This also hasn't happened, but one day it will, it'll hit the consumer market and that will be the death of "spinning-rus
Unfortunately all that is quite a myth for the most part.
Having worked in storage for a aeons the reality is that the difference between enterprise and "consumer grade rubbish" has very little to do anything but tollerance. If you picked up a 300G 10k enterprise drive and compared it to the consumer grade rubbish you'd find nothing different. It used to be the case, way back when, that they were very different but because consumer grade drives have gotten so much better its just not worth the expense of building the same drive for enterprise as for consumers with slightly different specs. What is different is the acceptable tollerances, when a platter comes off the line if its within 2% of its manufacturing tollerances its ok to use for entperise and if its higher they throw it into consumer. The reality is that most drives are in that "better than 2% tollerance" range and that is simply because the processes to make them have gotten so good over the years. The point is that when you hit your magic tollerance number, the drive is capable of 100% duty cycle.
So essentially, the difference between "consumer" and "enterprise" when it comes to the casing, the platters, the heads and the motors is zero. There are alot of different spec drives out there today ranging from 146gb (typically the smallest you'll find these days) all the way to 2gb with speeds form 7200 to 15000 rpm and enterprise is the only place that uses all of them, but they still come off the same manufacturing line. The drivers behind it all come down to the consumer itself, in enterprise its often about performance, and with consumers its about size. Very conveniently building bigger consumer grade drives typically means improving the performance of a drive in ways that scale straight back to the enterprise. Sure, you wont see many users throwing around 15k rpm drives, but thats more because its unnecessary.
So why is it that in the mid-to-low server range do we find 300gb 15k drives? Because its a cheap way of getting performance - and that is fairly important at that end of the market where servers need to be cheap and theres alot of competition (you know, 1-2ru with 4-8 drives and a raid card, no san).
So what else differs between the two? Interface. In the mid-to-low server range we start talking SAS and this is more to do with being able to talk to several drives at once (Again not something alot of consumers do other than with usb drives perhaps). The SAS interface is quite brilliant cause it can scale quite well to a larger number of drives than can SATA and does it very cheaply. It also takes alot of load off the server when it comes to processing data transfer (for a large number of drives). But in that same space you WILL find sata drives going up to 2tb (often servers lag consumers in size simply because of certification, not because of anything to do with stability). To call a 1tb drive unstable is rather silly in reality.
Now the BIG end of town - SAN's. These days in most SAN's you'll find a mix of SATA and Fibre channel (some do do SAS as well, but its uncommon though its changing). In the SAN end of town (the big boy game) you'll see it all. 7.2k rpm 2tb SATA's sitting in the same array along side 146g 15k RPM fibre channel and its all about trading off storage density/cost to performance. Consider this: 10 1tb sata drives can consume (easily) a 8gbps FC interface - OUCH! Now alot of SAN arrays start at around 4 FC intercaes and go up to maybe 16, but they'll be supporting literally thousands of drives. Alot of the SAN industry realised some time ago that throwing 2tb SATA's into an array made alot of sense because SAN interfaces have grown very slowly in terms of throughput and single HD interfaces have grown very quickly. There are even several very popular arrays that only do SATA and that was the driver behind "enterprise" grade large-storage drives (i.e. entperise grade 1tb+ sata drives). At the server you still get the fibre channel performance. The critical difference is that the array does more work
i got to book 9... god how i wish i could get those parts of my life back... by book 9 you'd read several hundred pages and think "what actually happened then?" and realise the whole series was becoming like days of our lives where only watching ever forth show was necessary (so im told, i do not watch days of our lives for the record).
Seriously talk about a waste of money, i cant believe i actually kept reading until book 9 - i am SUCK A DUMB A** (feel free to quote me on that in the future when i post annoying things).
unfortuantely im doing it alot, a hell of a lot actually...
So while my browser is getting flooded with these rediculous certs it wants to remember (or do i check the "dont remember permanently" box and deal with it even more often - i.e. more annoying) it slowly gets slower and slower because it has to record them all. Right now (just on my laptop) there are about 600 certs in that unknown catagory. Do you have any idea how much it speeds up my browser when i erase them all? Unfortunately, i do have some that I want to keep cause they're for various admin tasks and I cant change the cert. In several of our SAN storage gui's for example.
The best solution (for me) is to switch browser, which i will happily do when chrome is a little bit more stable. I can happily live without the plugins if i can get rid of that absolutely thoughtless ssl dialog.
And thats the point, I seriously cant believe they did this utterly stupid thing and didn't make it so people who work in IT and SI's in particular (where your going to stumble across a number of self-signed certs) can turn it off - theres a good reason they're the only ones who do it the way they do it.
Ever since ff 3, i have HATED firefox cause of its "invalid site certificate" rubbish. Do mozilla even get how absolutely annoying this is? Why oh god why would they do that. The way it was done previously (and the way EVERY OTHER BROWSER DOES IT was perfectly sufficient). Now, whenever I build a server (or a new network devide with a ssl-enabled web server for config) I have to go through that same pain. Given I work for an SI, this is frequent - and the reason I so desperately want to get rid of it in favour of chrome.
Now that chrome on ubuntu is getting might stable, i find myself using it alot more often. I've been dying for the day when chrome is good enough to replace firefox simply because mozilla dont give a rats bum about linux (well, thats not fair, they do but they could do alot better).
As for "windows 7 graphics features". Does anyone else get annoyed seeing what compiz (actually, it wasnt called compiz then, it was something else) was doing when aero came out simple because aero was crap in comparison? The reason I say that is cause i find the visual aspect of what linux is capable of quite impressive in comparison to windows (i.e. better) in alot of ways. All the while though I also hate where it seems to have fallen behind. Pull up nautilus for example - so much wasted white space (goes for all gnome apps really), yet the panels are beautifully efficient - i only have one. Same with ubuntu remix's window picker applet - nothing short of what a task bar should EXACTLY be, and its brilliant.
It was interesting as well to see google chrome's "how do we hack gnome/Qt to do that window bit like windows" - and they did it, which is great, but I do hope gnome get the message in terms of UI flexibility at the coding level. Time will tell.
I remember when i first got my hands on a sun thumper... impressive piece of kit but with a sun price tag... this one is way kewler on the price.
A project i worked on tried to deploy quite a number of thumper's and we ran into "issues"... First the racks - the thumper weighs in the vicinity of 150kgs (330lbs i think?), try put 10 of them in a rack and your in for a shock, assuming the rack can handle it, most data center floors have "issues" supporting the weight.
The second problem we had was cooling, the temp coming out the back of the rack was quite astronomical, and lastly power. In AU, this can often be a pain in the rear, specially with each thumper taking in about 2kw - 20kw per rack = PAIN.
Still, its kewl to see them do it all open.
I'd love to see someone do something like that though with computing power. Take the proliferation of mini-itx boards with "real" cpu's on them (ok, desktop cpus, but still, not shabby really) you could do a similar setup with a custom case supporting quite a number of those little buggers quite easily. Theres a beautiful little zotac AMD board that is almost ideal - supports the quad core, has a gig interface (sadly it has been aimed at htpc's cause you could replace the wireless, video, 6usb ports, etc with server-useful componentry - i.e. 2 or more gig ports and ipmi). But mostly it would be cheap and could run something like ovirt or abicloud quite happily. Shame that. There are other options in the space, like intel have a half-width xeon board (and a 1ru case that can support 2 of them side-by-side), but they're hard to get a hold of and quite long. get rid of local storage on the servers, use serial instead of video and add gpxe for remote boot - brilliant and dont exist!.
On a completely side note, one thing that i'd love to see in linux that has yet to exist in a useful format is replication (async) - the only real option is drbd, but its such a pain to setup and very inflexible. I was always so disappointed that neigther zfs, lvm or btrfs include it (even basic local replication would have surficed given the existence of so many network level block storage transports (iscsi, etc).
Oh, i forgot one of my main points... if google did drop X and its platform becomes popular, this could seriously damage alot of FOSS efforts out there (thats not to say someone might not come out with an X server for the OS), but consider some of our most important applications - firefox, openoffice, gimp, thunderbird, evolution... the list goes on... they all rely on a fairly comprehensive widget set google would be unlikely to provide if it was a non-X environment....
Admittedly, i've not following much about the chrome OS past the initial blog post but there are a few things about the original anouncement that bothered me (in a linux unfriendly way) and they all start with "... new window manager...".
The reason is simple, lets say you wanted to write an OS who's most important reason for existence is to be a web browser. Thats kinda exciting, but what could you could out of all this? X. Bear with me here. If you were a coder tasked with this little beauty, what would you be thinking? "I could do all this in chrome and get rid of x.org all together" especially if your targeting specific hardware platforms. The reason I think this is plausible is cause of google's own blog posts about the "nightmare" that is coding apps for linux distro's (i.e. multiple versions of this and that).
So what do you do? you take chrome, drop all the X/gtk/gnome code, add a little code that can talk directly to a video hardware interface in the kernel, add a little windowing code, add a dbus interface (think about that, that allows you to control almost all of the hardware on a linux machine), perhaps even add a little thing to talk to network manager seeing as how well that actually would control multiple types of networking interfaces these days. Add a small javascript routine to handle initial login stuff (for multiple users) and an audio control panel and we're done here, we have our web OS. The best part is? Its light, its fast, and it comes with none of the hideous chunk of code that is X/Fonts/Gnome/GTK/etc. Chrome already has a fairly basic file system browser - little effort to make it a bit more friendly and it'll function perfectly ok (you'll be doing some of this if your writing your own windowing stuff anyway).
The interesting one is games, something MS has dominated with no end in sight on the desktop and its the hardcore gamers that they all pitch to (EA, Blizzard, etc). Sure, take ipod style games and probably wouldnt be too hard to make them a web-deliverable (either a binary app coded for the google windowing platform or something that runs with "native client" or o3d or something), but hard core (you know, the ones that come on MULTIPLE dvd's these days?) are out of the question pretty much. Or perhaps they are not? remember a story a little while ago about pay-per-progression games? i.e. you get the game platform and it contains the basic content but you pay for new content as you progress through the game - that may lessen the pain of the multi-dvd games somewhat - again, speculation.
Also consider what apple did with the ipod/iphone. We have a system where by you cant even code for their platform without getting their SDK, if google were writing their own window manager, and its not X (by the way, thats my assumption, i've not seen anything to confirm that) and you could seriously do a very similar mechanism within chrome OS - i.e. a windowing platform thats licensed. I doubt google will go down that track, but there is an interesting tie-in with the way they handle android and its apps that could follow a similar path - i.e. SDK readily available - software distribution to handsets controlled by google. who's to say the jvm in android wouldn't make its way into chrome OS as well (Very likely really)...
just my 0.02c. In someway's i'd love to see it - i.e. an X replacement thats viable - but if your just aiming to do the bits above i mentioned, theres ALOT of things most window managers, systems and SDK do (i.e. widgets and stuff) and google just wouldn't need.... Then again, maybe we will all start coding web apps as desktop applications and tie it all into gears so you can run off-line as well... That wouldn't suck i dont think...
My personal opinion of apple has always been that it was a shame Microsoft were never quite able to land the final killing blow.
Microsoft are evil, but Apple hold the patent on it (for which MS pay licensing fee's).
Apple scare me more than MS ever has. Look at what apple has done to the iPhone for example - only software they approve? how is the DoJ not stepping in and saying "im sorry, thats a no-no?". The REALLY scarey thing is that if MS had of lost the look-and-feel debate of the 80's (or was it ealier?) we may very well have the same problem on the desktop (i.e. all running mac's with software only apple approve of).
Back on topic though, the pre cant sync via iTunes - its not the end of the world and i dont see it as apple being "evil" myself. Consider google aquiring licensing to sync with activesync for example, apple came with an idea they're not obliged to allow it to work with any other device. The reality is, if they were interested in interoperability they would have done it with syncml to begin with. But as we all know, both software and hardware vendors are interested in lock-in and open standards are the enemy.
Never forget that, OPEN STANDARDS ARE YOUR ENEMY IF YOU WORK IN THE HARDWARE AND SOFTWARE INDUSTRY. To think otherwise is a serious mis-calculation (with a few exceptions).
And that is why I almost exclusively use Linux - cause everyone who codes here at least partially aims to implement a standard (even if it can be quite painful sometimes). Granted, there are examples of that in reverse, but they are the exception rather then the norm. Even when linux does invent its own standards at least its all there in the source (and probably documentation) for you to implement. If you were ever after a reason for using linux, thats it.
i'll get off my soap-box now, i've got the flu and am on flu-drugs so i may be going off slightly half-cocked!
im not sure what corner of the planet your from, but the world i live in, the entire software industry has worked it REAR off to be as un-interoperable as possible. Heres a couple of examples that a number of people may even know (unless its me living on the other planet?)
Lets start with the "bios" common in most PC's today - go google that, i cant be bothered explaining several years of different pc architecture (bus, memory, cpu, hard disk connectors, all of it!) Microsoft Polute java - get sued and lose. TCP/IP - Microsoft go Netbios and Netbeui, Novel do IPX - both loose (thank god for that) Microsoft SMB/CIFS (and the various versions that went out of their way to break samba and the like), Appleshare, Novel NDS... Microsoft web "Standards" (or are they extensions? either way, the idea is to kill interop) Direct X (mostly direct 3d). Apple bios - no one boots anything but macos on an intel mac. J2EE - every vendor tries to do something to tie you to their own platform. Silverlight - another attempt at MS to control the web... ODF and OOXML - both made as broken as possible by microsoft (i.e. OOXML ISO != OOXML Office implementation, and MS Office ODF implementation not compatible with anyone elses).
This is one of those things that could go on and on and on when it comes to examples of how software companies have tried to be as anti-interop as possible... Now, seriously think back and see if you can name a few people who have tried (most in vane) to do the opposite and make computers interoperate - if you look closely, its a massive industry of its own!.
The fact that Linux, Windows and MacOS can even talk to each other over a network is mostly a miracle, though not hugely unsurprising.
Tom's hardware's blatant lying aside (12 disks + controller >> $1000). Its a good example of what people are getting sick to death of in the server market. i.e. add "enterprise" to a name and everything gets needlessly expensive.
if you look around at a couple of fibre-connect (or even scsi-connect) jbod's the cost is ludicrous even without the drives. Some of that is due to the weighty licensing costs of fibre channel and some of it is just greed (and an inability to efficiently produce components).
On the flip side, my personal experience with hardware raid controllers has never been great, the controller dies and you no longer have anyway of accessing your data cause only that controller knows how to read the meta data on your disks. Its also rather pointless in most situations these days. What exactly are you going to do with the much disk bandwidth? did you get a 10gb network suddenly you can share it over? About to do some film-quality HD video editing (in real time) for harry potter?
With the speed of most harddrives these days, theres just very little point using a hardware controller to build a "cheap system" when you could use a software one thats so much easier to replace if it fails. Your still going to get the speedy access.
I work for a reasonably small company, but i've also been asked by several of our clients (from small to mammoth) to help with the interview process for technical types. (Im in AU by the way, so may not be pertinent outside Australia).
The fact you may have come from a catastrophic failure really doesn't matter... There maybe some idle curiosity as to how it was "at the end", but in reality nothing matters as much as being able prove that:
1) you have the skill to do the job 2) you have the right attitude 3) other people are willing to say good things about you (i.e. references) and they are believable.
There's often very little you can assume/find out about a single persons involvement in a catastrophe. If they were there to the bitter end its either because they couldn't find a new job quick enough or because they were hoping to save the company (which are opposite ends of the spectrum in terms of an employee - but it could be a million other reasons too). The point im trying to make it that its impossible to know (from an employer perspective) what the reasons might have been and they rarely give much insight anyway. To some extent, the size of the company that folded can make it easier to figure out some facts, but its rarely a developers fault. Its often management/business decisions that result in the doom of a company.
All that aside, there are some things in the industry which are "more is better" and experience is probably the prime candidate. If you've been around for the end of a company, well thats something you have to your advantage as its an experience you have that perhaps alot of your possible competitors for a job may not. Simply put, as an employer if I had someone on my team that lost their job cause of a company collapse i'd certainly like to hear his views on where my company is heading (from a developers perspective).
Thats just my humble opinion, keep in mind almost all employers have different views on almost any subject. some may view your involvement in a catastrophe as a bad thing, but i personally don't know any.
(this is intended mostly as humor more than reality)
On the plus side, if any security group you buy software/hardware from gets hacked by these guys, you know that perhaps you choose the wrong security software/hardware provider... But, no doubt, the security consultant of their closest competitors will be knocking on your door shortly to sell their own product and show how anti-sec haven't hacked them yet!;)
Reading the text of their "manifesto" is quite interesting (assuming the link above actually points at what they said).
I don't believe its incredibly accurate (what they claim). Full-disclosure (if you've been around for a while) sort-of came about due to the security industries inability to actually respond to real threats (and they are still incapable of it). Often exploits would become available over the 'net from script-kiddie producers (i.e. the people with the real brains to figure out wholes in software and produce something even a script-kiddie could use) and so when something like SSH was "exploited" it was typically a case of the script kiddies being armed before the targets of the exploit.
Now-a-days, full disclosure mostly benefits the industry cause when the "ssh" attack came out, every person who wrote an ssh server could check to see if they were vulnerable and patch appropriately rather then say (only) f-secure finding out about the hack, fixing their own server software then running around telling everyone that "only we're secure!".
However, i dont get why imageshack were attacked, they seem to have very little to do with the people they claim they "are a target" of their rampage. Or was it just cause its such a widely used website that alot of people would see it where as most security-related sites are pretty low on the radar for alot of people?. What are imageshack doing running fedora core 5 (at least, the way i read that post they appear to be running an fc5 kernel)?
Of course being a linux advocate, why couldn't they have attacked a windows based server farm? Or made every ATM in the world print their message (now THAT would have gotten some serious publicity).
I always wanted to do an MMO based on a similar concept where you'd have some basic content for a fun game.. i.e. a single ship type flying through an infinitely large universe, then have users generate the rest of the content for you. It would be a very kewl concept, but its exceptionally hard to keep balanced. the kind of ideas i had though were along the lines of:
1) users generate content 2) users deploy content to test system 3) users vote on content (in or out) 4) content goes in, but sucks 5) modified content generated by users and deployed to test system 6) users vote on modified content 7) modified content goes into the game 8) code devlopers get some in-game reward for their efforts (i.e. 6 month patent on their design allowing only them to sell it in-game)
Now, it'd work best as FOSS because users could run their own servers and develop code onto it (plus, the community could be involved in the game right to the core - i.e. linux ports and so forth that companies seem to be really bad at), but i personally think that you'd need some kind of "we own the game" company to run it, partly to ensure continuous development, but also to avoid a constant battle of sharding servers (i.e. have a license that allows people to run servers, but not allow them to make money off them). I know that sounds very anti-gpl, but when it comes to things like that it would be hard to make it work in a way that doesnt end up with a tonne of small semi-thriving community serves (i.e. they voted my idea down, so im running my own server).
Then again, maybe it could become like irc networks (multiple servers run by multiple people connected together), and have say "Efnet" where some group of users has a set of functionality they like and "freenode" where they like different types of functionality.
I'd always wished EVE had adopted some strategy along the user generated content lines, i thought that was a perfect game for that type of thing (hell, the game is already based on python, developers would have no trouble picking it up and running with it)
When they say "new windowing system" it sounds suspiciously like "this aint X buddy". Which means alot of linux software (for the desktop) wouldn't run on it anyways.
on the plus side, it probably means that google will push hardware makers down in to the "create some kernel modules" route (even if its a route similar to the one nvidia have adopted).
The other thing that worries me is that google say things like "... need open source community..." and then bring us something like wave - "... need open source community help..." and i have yet to meet someone who is a FOSS dev who has access to the sandbox. So far its only been people at google or people sleeping with people at google. That was such a huge disappointment for someone who's a keen dev and was really impressed with a video... reminds me of the lovely vapourware concepts of not-so-long-ago
Slashdot Mirror
One thing that used to annoy me with "getting to my lag-free desktop" (and lets face it, thtas what its all about) was that i'd turn on and have to wait for a login prompt... then login and wait for a responsive desktop. When fed 9 came out I decided to use the luks encrypted disk - this turned out to be quite a massive bonus in some ways because the thing asks me for a password so early in the boot process.
I then turned on auto-login at the login screen and made firefox and evolution startup programs and wah lah my lifes (generally) alot simpler. i.e. turn on computer - 5 seconds later type in luks password - go make tea - come back to responsive desktop with my most used applications already open. A fast post is going to shave 5 seconds off my booting life? I really dont think im going to be that excited.
What I would prefer though is a rock-solid hibernate/restore strategy, and the idea (a little while ago) about a machine that shutdown, boots up and then hibernates does actually appeal but what kills that strategy is my laptop (where it would be useful) cause at home i have no docking station and at work I do and the thing can never hibernate between the two.
The only place i've ever found it useful is on the netbook where im booting up and down a fair bit and the only login is the encrypted disk login (right at the very beginning)... the rest of my machines I really dont care too much about fast boot times.
Actually, how often it boots and fast it boots are often small considerations - depending on where you sit.
Consider a file server that crashes and reboots twice a year and takes 5 minutes to come back up... thats 99.998% availability and from an infrastructure perspective thats pretty dang awesome.
From a helpdesk point of view, they'll suffer one heck of a beeting everytime it goes down and that'll be all they remember of the server.
Same goes for the users, all they'll remember is the 5 minutes it went down when half the company was doing something important.
Consider active directory though (or any kind of multi-master replicated service) - again, your talking about a server thats really not doing anything terribly difficult, but unlike the file server if it goes down no ones likely to notice. On top of that your often more concerned that the server will come up working then how quickly.
As for noise, well the good ol e450 from sun fully stacked with 20 15k disks (not to mention its in-built fans) used to make very amusing noises as it came up and some of the time it went down was because the disks managed to disconnect from the backplane (and you could hear the difference if you were around it during a boot often enough) - a good hard whack and it was sorted. Wondefull little machines those
Personally, on the server side, i remember the top end of sun's line back in its peek (from the e250, e450 all the way of to the e10k, the e25k, and the sunfire range)
5 minutes would have been a blessing on those machines.
Partly untrue, while the kind of benchmarks in the article have nothing (directly) to do with "Feel" alot of companies who have web sites (especially the big ones) do have tools to gauge feel.
The two i stumble across most often are from Segue and Mercury (Silk perform and load runner and their associated software bits). Seque's silk line is very good at judging feel, its designed to hit websites (as well as DB's and almost anything really) pretty hard with "real user access" patterns and can tell you how it'll perform from the user perspective depending on the browser they're using (it emulates various browsers). It can tell you things like the time taken to first start rendering the page, how long it took to render the page entirely and how long it took to download it all (as well as the impact of various link speeds). Those are the only two im particularly familiar with, but there are many many more. Ultimately, you can take all that data, and get a feel for "feel" based on browser type and link speed.
Coming form an SI type role, seeing what automated testing tools exist these days is pretty kewl (more used by testers then myself) and gives you a good feel of "when im building infrastructure what am I aiming to do".
I agree with your points for the most part... however, android has more in common with the PC then any other platform... plus, back in those days there were no FOSS alternatives (not real ones anyways)...
The reason i say that is similar to how people got into making PC - an open standard. Alot of phone makers would love one that allows them to make a phone that is fairly industry-wide compatible and the old real option at this point is windows.
Everyone would love to be able to build an iphone compatible but apple would never do that (Which is a pity imho cause that would catch on - you could see rim one day making that switch perhaps).
The big plus with android though is its potential for slimming down. Take the biggest owner in the market, nokia, the big sellers in their phones run on small chips that are just good enough to run symbian. Symbians greatest asset is that it can run on a much smaller footprint than windows can and ultimately makes for cheaper hardware. Android has a similar potential.
My point was simply this, android brings alot of potential to the industry, its not there yet, but it will be. consider:
1) its a sexy interface
2) its not exactly slow
3) its highly modifiable
4) it can be cut down to almost nothing
5) anyone can use it
6) its very powerful
(most of that I learnt by hacking my tytn2 to run android - horrible phone that tytn2, yet most people say its one of the best windows phones and probably sadly true)
which gives it all the advantages of all the other OS's and some more.
1) sexy like an iphone
2) slim like symbian
3) anyone can build on it like windows
4) its open like... err... openmoko? moblin?
5) its powerfull like iphone and windows mobile.
6) much like j2me phones, you dont have to target the cpu, but unlike j2me it has alot more power.
Anyways, thats my 0.02, its probably mostly wrong.
if im not mistaken, Wind River had two os's... VxWorks (Their original OS and had features like RTOS) and they also had something linuxy that they also had a vxworks brand on it..
I could be wrong though.
VxWorks was certianly nothing like linux though, so how you combine something destined for vxworks and linux together is beyond me... perhaps its jsut a badly worded statement or theres a back story here thats gone missing... maybe they're going to OSS VxWorks and roll up some of its components to linux? Or maybe they're going to make the two interoperable somehow? who knows?
And reading the actual article I can entirely agree with him, but I also think its partly his own fault.
Linus himself (not that long ago) was talking about how they "deliberately" make the kernel code difficult to understand and he had good reasons for that which revolved around not getting people interested unless they're willing to stick with it... i.e. to avoid orphaned code. From my own experience I started writing a kernel module in the scsi stack that i later abandoned because another piece of work expanded to provide the exact same functionality.
The problem though, for me, was that the entire time I was writing my little module it kept occuring to me that there were other parts of the stack I could easily hand off to rather then re-writing it myself. Unfortunately the code is often so hard to read that you spend more time trying to understand if some other segment of kernel code is actually doing what you need to do. Thats painful. Ultimately you end up writing your own little bit of code because you need to be intimately familiar with other branches of code to understand whether you need to implement something or not.
Hence the bloat, im sure there is tonnes of code that could be moved into higher levels of functionality in the kernel that instead get written over and over again even though they may be able to be reused. An example is SCSI and Network, now they are very different branches of code dealing with totally disparate systems but ultimately they do similar things, i.e. take a chunk of data from one interface and dump into another interface. I'd be curious to know how much of what they do gets passed into some higher level kernel code though...?
Raid may be coming to end, but its nothing to do with the speeds of drives. Its all about size. Its all about the fact that you are GUARANTEED to have a sector fail on read after x number of reads. Now because x hasnt changed much over the decades while drives have gotten vastly larger, we're approaching x but we've still got a while to go yet and in the enterprise (where raid actually matters) its not something that'll really be a problem for at least another 5-10 years. This is not a problem that raid 6 or zfs solve.
But, the problem is this, if a drive fails in a raid 5 array and the array runs off to rebuild it then when you get near x you'll fail during the rebuild and that will keep occuring until the array gives up. Raid 6 still has the same issue, it'll still read all the drives to rebuild the broken disk, same with ZFS. Unfortunately, there has been alot of industry spin suggesting raid 6 and ZFS solve this issue which is patently false.
Raid 6 was designed because of how arrays were being deployed, when large scale disk boxes were being deployed (SAN etc) we got to a point where if you had enough disks you might as well have an engineer on site because the failure rate of drives was such that when you got to a certain number of drives (we're talking hundred of thousands here, not that hard in a large data center packed with SAN arrays) you could guarentee at least one drive would fail EVERY SINGLE DAY. So, with raid sets getting larger (i.e. 10-15 drives in a single raid 5 array and they were getting much bigger than that) it became important to save yourself from a dual-drive failure (because the likelyhood of it happening went up exponentially as the set got larger), having two parity drives solved that issue - hence raid 6.
ZFS was designed for performance. Raid 5 (and 6) SUCKS for performance when your not using some form of off-load device (hardware raid card or SAN array head) and so ZFS was designed to (in part) give people the ability to do that kind of thing on the server cpu without killing the performance of having 6-10 striped drives. The problem is that if you throw some data at a drive, computers are very good at it. They can just say "write this chunk of data here please" and the rest just happens. With raid (if its done in software), what you have to do is calculate the parity of ever single byte of data on the array - this hurts alot and destroys the whole transfer mechanism.
The point is though, we're approaching x and the only real solution is to fix x. Though, with SSD's coming along quite nicely, the problem will likely resolve itself and we'll be using raid 6 SSD's. SSD's are quite capable of solving this little drama because x can be quite variable unlike their "spinning-rusty-metal" counterparts. They can also be scaled alot easier for physical size (something we havent seen yet). Its very hard to produce a tonne of different sizes when it comes to disks - i.e. we have three formats, 1.8", 2.5" and 3.25" and alot of that has to do with producing some "spinning-rusty-metal" in an efficient way at different sizes. i.e. thats just not practical for motorized components. SSD's on the other hand are a completely different boat. Because they are just chips on a little board they can scale very well in terms of physical size. Right now it hasnt happened, they are doomed to follow the pre-set sizes of their cousins because of the fact we're forced to use both.
This wont always be the case, eventually not only will SATA, FC, infiniband (an already mostly dead tech any way) and SAS be a thing of the past (They were designed for spinning rusty metal and with people like seagate threatening to sue ssd makers for using sata and the like, ssd boys will probably make their own interface). The traditional interfaces for HD's are not fantastic (as it turns out) for ssd's, you could in fact make a much better interface if your talking directly to ssd's. This also hasn't happened, but one day it will, it'll hit the consumer market and that will be the death of "spinning-rus
Unfortunately all that is quite a myth for the most part.
Having worked in storage for a aeons the reality is that the difference between enterprise and "consumer grade rubbish" has very little to do anything but tollerance. If you picked up a 300G 10k enterprise drive and compared it to the consumer grade rubbish you'd find nothing different. It used to be the case, way back when, that they were very different but because consumer grade drives have gotten so much better its just not worth the expense of building the same drive for enterprise as for consumers with slightly different specs. What is different is the acceptable tollerances, when a platter comes off the line if its within 2% of its manufacturing tollerances its ok to use for entperise and if its higher they throw it into consumer. The reality is that most drives are in that "better than 2% tollerance" range and that is simply because the processes to make them have gotten so good over the years. The point is that when you hit your magic tollerance number, the drive is capable of 100% duty cycle.
So essentially, the difference between "consumer" and "enterprise" when it comes to the casing, the platters, the heads and the motors is zero. There are alot of different spec drives out there today ranging from 146gb (typically the smallest you'll find these days) all the way to 2gb with speeds form 7200 to 15000 rpm and enterprise is the only place that uses all of them, but they still come off the same manufacturing line. The drivers behind it all come down to the consumer itself, in enterprise its often about performance, and with consumers its about size. Very conveniently building bigger consumer grade drives typically means improving the performance of a drive in ways that scale straight back to the enterprise. Sure, you wont see many users throwing around 15k rpm drives, but thats more because its unnecessary.
So why is it that in the mid-to-low server range do we find 300gb 15k drives? Because its a cheap way of getting performance - and that is fairly important at that end of the market where servers need to be cheap and theres alot of competition (you know, 1-2ru with 4-8 drives and a raid card, no san).
So what else differs between the two? Interface. In the mid-to-low server range we start talking SAS and this is more to do with being able to talk to several drives at once (Again not something alot of consumers do other than with usb drives perhaps). The SAS interface is quite brilliant cause it can scale quite well to a larger number of drives than can SATA and does it very cheaply. It also takes alot of load off the server when it comes to processing data transfer (for a large number of drives). But in that same space you WILL find sata drives going up to 2tb (often servers lag consumers in size simply because of certification, not because of anything to do with stability). To call a 1tb drive unstable is rather silly in reality.
Now the BIG end of town - SAN's. These days in most SAN's you'll find a mix of SATA and Fibre channel (some do do SAS as well, but its uncommon though its changing). In the SAN end of town (the big boy game) you'll see it all. 7.2k rpm 2tb SATA's sitting in the same array along side 146g 15k RPM fibre channel and its all about trading off storage density/cost to performance. Consider this: 10 1tb sata drives can consume (easily) a 8gbps FC interface - OUCH! Now alot of SAN arrays start at around 4 FC intercaes and go up to maybe 16, but they'll be supporting literally thousands of drives. Alot of the SAN industry realised some time ago that throwing 2tb SATA's into an array made alot of sense because SAN interfaces have grown very slowly in terms of throughput and single HD interfaces have grown very quickly. There are even several very popular arrays that only do SATA and that was the driver behind "enterprise" grade large-storage drives (i.e. entperise grade 1tb+ sata drives). At the server you still get the fibre channel performance. The critical difference is that the array does more work
i got to book 9... god how i wish i could get those parts of my life back... by book 9 you'd read several hundred pages and think "what actually happened then?" and realise the whole series was becoming like days of our lives where only watching ever forth show was necessary (so im told, i do not watch days of our lives for the record).
Seriously talk about a waste of money, i cant believe i actually kept reading until book 9 - i am SUCK A DUMB A** (feel free to quote me on that in the future when i post annoying things).
unfortuantely im doing it alot, a hell of a lot actually...
So while my browser is getting flooded with these rediculous certs it wants to remember (or do i check the "dont remember permanently" box and deal with it even more often - i.e. more annoying) it slowly gets slower and slower because it has to record them all. Right now (just on my laptop) there are about 600 certs in that unknown catagory. Do you have any idea how much it speeds up my browser when i erase them all? Unfortunately, i do have some that I want to keep cause they're for various admin tasks and I cant change the cert. In several of our SAN storage gui's for example.
The best solution (for me) is to switch browser, which i will happily do when chrome is a little bit more stable. I can happily live without the plugins if i can get rid of that absolutely thoughtless ssl dialog.
And thats the point, I seriously cant believe they did this utterly stupid thing and didn't make it so people who work in IT and SI's in particular (where your going to stumble across a number of self-signed certs) can turn it off - theres a good reason they're the only ones who do it the way they do it.
Ever since ff 3, i have HATED firefox cause of its "invalid site certificate" rubbish. Do mozilla even get how absolutely annoying this is? Why oh god why would they do that. The way it was done previously (and the way EVERY OTHER BROWSER DOES IT was perfectly sufficient). Now, whenever I build a server (or a new network devide with a ssl-enabled web server for config) I have to go through that same pain. Given I work for an SI, this is frequent - and the reason I so desperately want to get rid of it in favour of chrome.
Now that chrome on ubuntu is getting might stable, i find myself using it alot more often. I've been dying for the day when chrome is good enough to replace firefox simply because mozilla dont give a rats bum about linux (well, thats not fair, they do but they could do alot better).
As for "windows 7 graphics features". Does anyone else get annoyed seeing what compiz (actually, it wasnt called compiz then, it was something else) was doing when aero came out simple because aero was crap in comparison? The reason I say that is cause i find the visual aspect of what linux is capable of quite impressive in comparison to windows (i.e. better) in alot of ways. All the while though I also hate where it seems to have fallen behind. Pull up nautilus for example - so much wasted white space (goes for all gnome apps really), yet the panels are beautifully efficient - i only have one. Same with ubuntu remix's window picker applet - nothing short of what a task bar should EXACTLY be, and its brilliant.
It was interesting as well to see google chrome's "how do we hack gnome/Qt to do that window bit like windows" - and they did it, which is great, but I do hope gnome get the message in terms of UI flexibility at the coding level. Time will tell.
I remember when i first got my hands on a sun thumper... impressive piece of kit but with a sun price tag... this one is way kewler on the price.
A project i worked on tried to deploy quite a number of thumper's and we ran into "issues"... First the racks - the thumper weighs in the vicinity of 150kgs (330lbs i think?), try put 10 of them in a rack and your in for a shock, assuming the rack can handle it, most data center floors have "issues" supporting the weight.
The second problem we had was cooling, the temp coming out the back of the rack was quite astronomical, and lastly power. In AU, this can often be a pain in the rear, specially with each thumper taking in about 2kw - 20kw per rack = PAIN.
Still, its kewl to see them do it all open.
I'd love to see someone do something like that though with computing power. Take the proliferation of mini-itx boards with "real" cpu's on them (ok, desktop cpus, but still, not shabby really) you could do a similar setup with a custom case supporting quite a number of those little buggers quite easily. Theres a beautiful little zotac AMD board that is almost ideal - supports the quad core, has a gig interface (sadly it has been aimed at htpc's cause you could replace the wireless, video, 6usb ports, etc with server-useful componentry - i.e. 2 or more gig ports and ipmi). But mostly it would be cheap and could run something like ovirt or abicloud quite happily. Shame that. There are other options in the space, like intel have a half-width xeon board (and a 1ru case that can support 2 of them side-by-side), but they're hard to get a hold of and quite long. get rid of local storage on the servers, use serial instead of video and add gpxe for remote boot - brilliant and dont exist!.
On a completely side note, one thing that i'd love to see in linux that has yet to exist in a useful format is replication (async) - the only real option is drbd, but its such a pain to setup and very inflexible. I was always so disappointed that neigther zfs, lvm or btrfs include it (even basic local replication would have surficed given the existence of so many network level block storage transports (iscsi, etc).
Oh, i forgot one of my main points... if google did drop X and its platform becomes popular, this could seriously damage alot of FOSS efforts out there (thats not to say someone might not come out with an X server for the OS), but consider some of our most important applications - firefox, openoffice, gimp, thunderbird, evolution... the list goes on... they all rely on a fairly comprehensive widget set google would be unlikely to provide if it was a non-X environment....
Admittedly, i've not following much about the chrome OS past the initial blog post but there are a few things about the original anouncement that bothered me (in a linux unfriendly way) and they all start with "... new window manager ...".
The reason is simple, lets say you wanted to write an OS who's most important reason for existence is to be a web browser. Thats kinda exciting, but what could you could out of all this? X. Bear with me here. If you were a coder tasked with this little beauty, what would you be thinking? "I could do all this in chrome and get rid of x.org all together" especially if your targeting specific hardware platforms. The reason I think this is plausible is cause of google's own blog posts about the "nightmare" that is coding apps for linux distro's (i.e. multiple versions of this and that).
So what do you do? you take chrome, drop all the X/gtk/gnome code, add a little code that can talk directly to a video hardware interface in the kernel, add a little windowing code, add a dbus interface (think about that, that allows you to control almost all of the hardware on a linux machine), perhaps even add a little thing to talk to network manager seeing as how well that actually would control multiple types of networking interfaces these days. Add a small javascript routine to handle initial login stuff (for multiple users) and an audio control panel and we're done here, we have our web OS. The best part is? Its light, its fast, and it comes with none of the hideous chunk of code that is X/Fonts/Gnome/GTK/etc. Chrome already has a fairly basic file system browser - little effort to make it a bit more friendly and it'll function perfectly ok (you'll be doing some of this if your writing your own windowing stuff anyway).
The interesting one is games, something MS has dominated with no end in sight on the desktop and its the hardcore gamers that they all pitch to (EA, Blizzard, etc). Sure, take ipod style games and probably wouldnt be too hard to make them a web-deliverable (either a binary app coded for the google windowing platform or something that runs with "native client" or o3d or something), but hard core (you know, the ones that come on MULTIPLE dvd's these days?) are out of the question pretty much. Or perhaps they are not? remember a story a little while ago about pay-per-progression games? i.e. you get the game platform and it contains the basic content but you pay for new content as you progress through the game - that may lessen the pain of the multi-dvd games somewhat - again, speculation.
Also consider what apple did with the ipod/iphone. We have a system where by you cant even code for their platform without getting their SDK, if google were writing their own window manager, and its not X (by the way, thats my assumption, i've not seen anything to confirm that) and you could seriously do a very similar mechanism within chrome OS - i.e. a windowing platform thats licensed. I doubt google will go down that track, but there is an interesting tie-in with the way they handle android and its apps that could follow a similar path - i.e. SDK readily available - software distribution to handsets controlled by google. who's to say the jvm in android wouldn't make its way into chrome OS as well (Very likely really)...
just my 0.02c. In someway's i'd love to see it - i.e. an X replacement thats viable - but if your just aiming to do the bits above i mentioned, theres ALOT of things most window managers, systems and SDK do (i.e. widgets and stuff) and google just wouldn't need.... Then again, maybe we will all start coding web apps as desktop applications and tie it all into gears so you can run off-line as well... That wouldn't suck i dont think...
Only time will tell where it goes...
My personal opinion of apple has always been that it was a shame Microsoft were never quite able to land the final killing blow.
Microsoft are evil, but Apple hold the patent on it (for which MS pay licensing fee's).
Apple scare me more than MS ever has. Look at what apple has done to the iPhone for example - only software they approve? how is the DoJ not stepping in and saying "im sorry, thats a no-no?". The REALLY scarey thing is that if MS had of lost the look-and-feel debate of the 80's (or was it ealier?) we may very well have the same problem on the desktop (i.e. all running mac's with software only apple approve of).
Back on topic though, the pre cant sync via iTunes - its not the end of the world and i dont see it as apple being "evil" myself. Consider google aquiring licensing to sync with activesync for example, apple came with an idea they're not obliged to allow it to work with any other device. The reality is, if they were interested in interoperability they would have done it with syncml to begin with. But as we all know, both software and hardware vendors are interested in lock-in and open standards are the enemy.
Never forget that, OPEN STANDARDS ARE YOUR ENEMY IF YOU WORK IN THE HARDWARE AND SOFTWARE INDUSTRY. To think otherwise is a serious mis-calculation (with a few exceptions).
And that is why I almost exclusively use Linux - cause everyone who codes here at least partially aims to implement a standard (even if it can be quite painful sometimes). Granted, there are examples of that in reverse, but they are the exception rather then the norm. Even when linux does invent its own standards at least its all there in the source (and probably documentation) for you to implement. If you were ever after a reason for using linux, thats it.
i'll get off my soap-box now, i've got the flu and am on flu-drugs so i may be going off slightly half-cocked!
im not sure what corner of the planet your from, but the world i live in, the entire software industry has worked it REAR off to be as un-interoperable as possible. Heres a couple of examples that a number of people may even know (unless its me living on the other planet?)
Lets start with the "bios" common in most PC's today - go google that, i cant be bothered explaining
several years of different pc architecture (bus, memory, cpu, hard disk connectors, all of it!)
Microsoft Polute java - get sued and lose.
TCP/IP - Microsoft go Netbios and Netbeui, Novel do IPX - both loose (thank god for that)
Microsoft SMB/CIFS (and the various versions that went out of their way to break samba and the like), Appleshare, Novel NDS...
Microsoft web "Standards" (or are they extensions? either way, the idea is to kill interop)
Direct X (mostly direct 3d).
Apple bios - no one boots anything but macos on an intel mac.
J2EE - every vendor tries to do something to tie you to their own platform.
Silverlight - another attempt at MS to control the web...
ODF and OOXML - both made as broken as possible by microsoft (i.e. OOXML ISO != OOXML Office implementation, and MS Office ODF implementation not compatible with anyone elses).
This is one of those things that could go on and on and on when it comes to examples of how software companies have tried to be as anti-interop as possible... Now, seriously think back and see if you can name a few people who have tried (most in vane) to do the opposite and make computers interoperate - if you look closely, its a massive industry of its own!.
The fact that Linux, Windows and MacOS can even talk to each other over a network is mostly a miracle, though not hugely unsurprising.
Tom's hardware's blatant lying aside (12 disks + controller >> $1000). Its a good example of what people are getting sick to death of in the server market. i.e. add "enterprise" to a name and everything gets needlessly expensive.
if you look around at a couple of fibre-connect (or even scsi-connect) jbod's the cost is ludicrous even without the drives. Some of that is due to the weighty licensing costs of fibre channel and some of it is just greed (and an inability to efficiently produce components).
On the flip side, my personal experience with hardware raid controllers has never been great, the controller dies and you no longer have anyway of accessing your data cause only that controller knows how to read the meta data on your disks. Its also rather pointless in most situations these days. What exactly are you going to do with the much disk bandwidth? did you get a 10gb network suddenly you can share it over? About to do some film-quality HD video editing (in real time) for harry potter?
With the speed of most harddrives these days, theres just very little point using a hardware controller to build a "cheap system" when you could use a software one thats so much easier to replace if it fails. Your still going to get the speedy access.
I work for a reasonably small company, but i've also been asked by several of our clients (from small to mammoth) to help with the interview process for technical types. (Im in AU by the way, so may not be pertinent outside Australia).
The fact you may have come from a catastrophic failure really doesn't matter... There maybe some idle curiosity as to how it was "at the end", but in reality nothing matters as much as being able prove that:
1) you have the skill to do the job
2) you have the right attitude
3) other people are willing to say good things about you (i.e. references) and they are believable.
There's often very little you can assume/find out about a single persons involvement in a catastrophe. If they were there to the bitter end its either because they couldn't find a new job quick enough or because they were hoping to save the company (which are opposite ends of the spectrum in terms of an employee - but it could be a million other reasons too). The point im trying to make it that its impossible to know (from an employer perspective) what the reasons might have been and they rarely give much insight anyway. To some extent, the size of the company that folded can make it easier to figure out some facts, but its rarely a developers fault. Its often management/business decisions that result in the doom of a company.
All that aside, there are some things in the industry which are "more is better" and experience is probably the prime candidate. If you've been around for the end of a company, well thats something you have to your advantage as its an experience you have that perhaps alot of your possible competitors for a job may not. Simply put, as an employer if I had someone on my team that lost their job cause of a company collapse i'd certainly like to hear his views on where my company is heading (from a developers perspective).
Thats just my humble opinion, keep in mind almost all employers have different views on almost any subject. some may view your involvement in a catastrophe as a bad thing, but i personally don't know any.
(this is intended mostly as humor more than reality)
On the plus side, if any security group you buy software/hardware from gets hacked by these guys, you know that perhaps you choose the wrong security software/hardware provider... But, no doubt, the security consultant of their closest competitors will be knocking on your door shortly to sell their own product and show how anti-sec haven't hacked them yet! ;)
Reading the text of their "manifesto" is quite interesting (assuming the link above actually points at what they said).
I don't believe its incredibly accurate (what they claim). Full-disclosure (if you've been around for a while) sort-of came about due to the security industries inability to actually respond to real threats (and they are still incapable of it). Often exploits would become available over the 'net from script-kiddie producers (i.e. the people with the real brains to figure out wholes in software and produce something even a script-kiddie could use) and so when something like SSH was "exploited" it was typically a case of the script kiddies being armed before the targets of the exploit.
Now-a-days, full disclosure mostly benefits the industry cause when the "ssh" attack came out, every person who wrote an ssh server could check to see if they were vulnerable and patch appropriately rather then say (only) f-secure finding out about the hack, fixing their own server software then running around telling everyone that "only we're secure!".
However, i dont get why imageshack were attacked, they seem to have very little to do with the people they claim they "are a target" of their rampage. Or was it just cause its such a widely used website that alot of people would see it where as most security-related sites are pretty low on the radar for alot of people?. What are imageshack doing running fedora core 5 (at least, the way i read that post they appear to be running an fc5 kernel)?
Of course being a linux advocate, why couldn't they have attacked a windows based server farm? Or made every ATM in the world print their message (now THAT would have gotten some serious publicity).
I always wanted to do an MMO based on a similar concept where you'd have some basic content for a fun game.. i.e. a single ship type flying through an infinitely large universe, then have users generate the rest of the content for you. It would be a very kewl concept, but its exceptionally hard to keep balanced. the kind of ideas i had though were along the lines of:
1) users generate content
2) users deploy content to test system
3) users vote on content (in or out)
4) content goes in, but sucks
5) modified content generated by users and deployed to test system
6) users vote on modified content
7) modified content goes into the game
8) code devlopers get some in-game reward for their efforts (i.e. 6 month patent on their design allowing only them to sell it in-game)
Now, it'd work best as FOSS because users could run their own servers and develop code onto it (plus, the community could be involved in the game right to the core - i.e. linux ports and so forth that companies seem to be really bad at), but i personally think that you'd need some kind of "we own the game" company to run it, partly to ensure continuous development, but also to avoid a constant battle of sharding servers (i.e. have a license that allows people to run servers, but not allow them to make money off them). I know that sounds very anti-gpl, but when it comes to things like that it would be hard to make it work in a way that doesnt end up with a tonne of small semi-thriving community serves (i.e. they voted my idea down, so im running my own server).
Then again, maybe it could become like irc networks (multiple servers run by multiple people connected together), and have say "Efnet" where some group of users has a set of functionality they like and "freenode" where they like different types of functionality.
I'd always wished EVE had adopted some strategy along the user generated content lines, i thought that was a perfect game for that type of thing (hell, the game is already based on python, developers would have no trouble picking it up and running with it)
When they say "new windowing system" it sounds suspiciously like "this aint X buddy". Which means alot of linux software (for the desktop) wouldn't run on it anyways.
on the plus side, it probably means that google will push hardware makers down in to the "create some kernel modules" route (even if its a route similar to the one nvidia have adopted).
Interestingly, i blogged about an OS I thought would be perfect for google (also linux based) http://pjrlost.blogspot.com/2009/07/desktop-os-that-google-could-do.html that was more along the lines of "consistent user experience" and a fully fledged OS.
The other thing that worries me is that google say things like "... need open source community ..." and then bring us something like wave - "... need open source community help ..." and i have yet to meet someone who is a FOSS dev who has access to the sandbox. So far its only been people at google or people sleeping with people at google. That was such a huge disappointment for someone who's a keen dev and was really impressed with a video... reminds me of the lovely vapourware concepts of not-so-long-ago