Answer: network [apple.com] and software [apple.com] compatibility, of course, which Apple is clearly working very hard on.
Network and file format compatability are lemons, especially considering the open source crew have done 90% of the work for them. Software compatability is what matters, being able to run custom apps, and of course all that other software that exists out there on the net that uses the Win32 GUI.
A very valid point. I blame the morons who didn't know how to write portable code, as well as the morons who didn't require it in the first place. Yes, the Win32 GUI isn't going to port, but if the app was well written, porting it would be a viable option.
Have you ever tried to write portable software of any complexity at all? If you think it's merely a case of saying "oh, well, I'll make this software totally cross platform" then you need to get more experience of programming in the real world. Writing cross platform software is hard with a capital H. Notice how much duplication there is in most portable projects? Mozilla and OpenOffice both implement their own widget toolkits, portability libraries, object models etc. The GIMP relies upon a dodgy windows port of GTK. The fact is, that writing cross platform software that is high quality on all platforms is a nightmare, and for custom business apps, where ALL your computers are Windows, needlessly spending money on portability would make you uncompetitive. So all your competitors are using software X to help them get ahead, but you're not going to because it won't run on an OS you don't use? Sounds like a going out of business strategy to me. The only solution is something like Wine.
So what if the Mac has no equivalent, as long as it can connect to the same CrossOver Office server?
So your solution to running Windows software on the Mac is to load it up with an X server and use a Linux box as well? That could work for Office. That assumes a lot though, like all software being multi-instance. Most of it isn't, because doing all the locking etc of data files needed is also hard and for desktop apps on Windows, historically single user, 99% of the time not needed. Office can do it with a couple of hacks, because they needed it for Citrix originally afaik. Most can't, you need to run a copy of each desktop to get good performance, if it runs at all.
So the question is, what on earth will compell them to drop Windows on the desktop? Because it's sure as hell not any of the issues we've seen so far.
They will drop Windows when there is an alternative that is 200% as good as Windows, 100% compatible, and only 10% of the cost. Oh yeah, it must have excellent online training too.
Linux hasn't reached all those figures yet (though it's well on the way), and has no online training of any merit (or indeed online help all too often). When it has all those, then companies will switch.
Hmmm, perhaps, but Apple sell hardware primarily, and there's nothing wrong with the hardware most business has - the issue is with software.
A medium sized company can easily have 5000 desktops. The average price of a Mac is I'd guess about £1000, so that's a cool £5,000,000 (about $7,500,000) just to replace hardware that already worked? You might as well pay the fees to Microsoft, that's almost certainly cheaper. And don't forget that most businesses have at least 1 or 2 custom apps.
The obvious solution is Linux - with a decent set of administators Linux is within a year of being just great on the corporate desktop. The final usability problems are being hammered at a truly astonishing rate, and with tricks like CrossOver Office Server you can pay for 1 copy of Office (I guess it'd work with other programs too) but have it serve hundreds of desktops. Wine is so critical in these areas, for custom business apps, and the Mac has no equivalent, probably won't for some time, if ever.
Read the whitepaper on namesys.com - he'll be introduced set theoretic naming later on, allowing for associative databases (not relational like longhorn).
Use the list view, and navagate with the arrow keys (or by typing the name of a file) and pressing Apple-O. It's like a graphical command-line that way.
Yes, I think Explorer can do something similar. It's not quite the same, but as I'm not a heavy Finder user I'll have to take your word for it....
The Finder doesn't take up memory with caching all the directory structures, just the most recent. So the rendering is slow due to the Finder reading the icons + positions + the directory listing and all the other info from the disk.
Hmm? Neither does Rox, but it's still very fast. I don't know how Tom Leonard does it, but I recall seeing a brief explanation. It's got some pretty clever internals to make it extremely fast - which is why people associate rox with fast. "Buy a faster disk" isn't really a good solution compared to "make the finder faster".
Ahh, now we're getting somewhere. This has been a problem since OS X, because type and creator codes are no longer required, and thus files dont' always have them. The most common types such as.png,.jpg, etc. are supposed to be noticed by the finder and passed to QuickTime for this same sort of detection, all behind the scenes. [snip]
Hmm, sounds like it was good, albiet very complex. 10/10 for flexibility. Wouldn't running apps themselves have quite a bit of overhead though if you wanted to compile say a view with metadata in it? file is very fast as it just uses a combination of lookup tables with binary regexs iirc.
A simple RTFM solves this. Use the technique I described above, type the name until the highlight matches, and Apple-O, or any other keyboard shortcut you'd like, depending of course, on what you want to do with the file. You can actually move quite fast once you get the hang of it, it comes as natural as typing and using the shift key.
What happens if I have two files called "Business Plan" and "Business Diary" (let's pretend). Can I type "bu - tab - d - enter"? Or do I have to type "Business D - enter"? This isn't trying to trip you up or anything, I'm just curious.
Ahh, yes this feature coudl stand out a little more but the Finder is no Web Browser and therefore we keep this tucked away under the Go menu. Choose "Connect To Server..." or, for speed-shortcutting, use Apple-K (Konnect, C interferes with Copy)
OK, cool. Is that a VFS type system? ie in Konqueror I can go to audiocd:/ and copy my audio tracks named from a CDDB server as Oggs to my hard disk, and the files will be ripped and compressed transparently. There's nothing magic about that, what's neat is that it looks just like another FS to me. You could argue this is bloat, but I see no reason why it should be slow technically....
You clearly haven't used the Finder much, IamTheRealMike, but then you also seem to not care because it is Apple. This is a theme that came across in your post. Dont' assume you need to tell someone off because they defended their view - if I was to bad mouth the Linux kernel I'd be killed on this website, by all the slashdotters doing the same. Give it another try, look around carefully, Apple does things differently for a reason. You won't be a power user in a day. That's why Apple's interface is great, it does things logically but not the way other interfaces too. There is too much similarity with other file borwsers that limits them in some respects.
Well, good points. I wrote that last paragraph mainly because toupsie is a well known Mac troll, pretty much every post I see from him/her that gets modded up is estolling the Mac and trashing everything else. I'm not a heavy Finder user, I tend to use it on other peoples Macs, and I don't actually hate the Finder or anything, but I was trying to counterbalance toupsie who seemed to think that the Finder was perfection (it isn't) and that Explorer/the CLI/Nautilus/whatever is automatically awful, because it isn't Apple (they aren't)
Well, thanks for the informative reply. I wish all replies to my posts were like that......
Re:Don't compare Mac OS Finder to Windows Explorer
on
The Captains of Nautilus
·
· Score: 5, Informative
I would read the Human Interface Guidlines as mentioned in other replies. Next I would get my hands on a Mac and play with it. I hate to say it, but Finder is one of those things you need to experience to see the difference.
You're enthusiasm for Apples products is great, but I feel I give an alternative viewpoint. I've used the Finder, Explorer, Nautilus, Konqueror, Rox, and the command line, so I have quite a bit to compare it to.
I feel the Finder sucks bigtime:
The NeXT style columns view (the default) is awful. I found I couldn't get it to display as much info as Explorer could in the same space, I found that copying between two locations meant I had to open 2 finder windows or engage the rather feeble tree widget. It wastes space, the big icon/preview is very pretty, but 90% of the time useless as I already know what the filetype is, it just takes up a big fat wad of space that could have been used for something else.
It's slow. No really, even on 10.2, I could watch as it rerendered the Finder on a complex directory structure. Quartz Extreme me all you like, I didn't try it with that, as the drivers for the card in the machine I was using didn't support it (the owner had upgraded it himself). Rox is fast. The Finder is slow.
Primitive typing: in Rox if I view the properties of a file, the "file" program will scan it and try to figure out what exactly it is. It'll say for instance "Screenshot.png: PNG image data, 1024 x 768, 8-bit/color RGB, non-interlaced" or "ASCII Text, long lines". The Finder just says "Document" for any type that isn't explicitly registered with it, at least rox tries to guess based on some reasonably smart heuristics
Apparently no Rox/Nautilus type-ahead tab complete. This isn't a "hard" feature, once you know it's there anybody can use it, I've seen die hard Windowsers pick it up in less than 10 seconds. If the Finder has it, it didn't make it particularly obvious. Rox has a great implementation, just hit / and use it like you would the CLI, you can see it scan through the directories as you type, and get visual feedback as it matches. Nautilus2 has something similar though not as slick if you press Ctrl-L
No address bar? I feel sure it can mount FTP drives etc as it can do the iDisk, but there's no obvious place to type in any URLs for that.
You clearly like the Finder toupsie, but then you like anything that is Apple, and hate anything that isn't, this is a theme that comes across in most of your posts. File management is very much a personal thing - don't assume your view is the "right" view. Comments like "I can't explain, you must just use it" don't help your arguments by the way.
Yes, but there is nothing wrong in borrowing an idea as long as it is a good idea
But it is a good idea? I'd bet 90% of Linux users are ex-windows users, which by definition meant ex-explorer users. Yet most Linux users I know use the command line. Perhaps because once you've learnt it, it's faster and more efficient?
Yes, but I can bet that an application like Nautilus is necessary if you ever want to have a decent market share on the desktop . By decent I mean 10% or more.
I think it's a shame we feel we have to copy Windows in order to have Real People (tm) use it. Remember, there is no geek/real person divide, only shades in between. As Linux gets easier to use, so more people overcome the learning curve and the community of users grows. Some people will never, ever want to do anything in a way that's different to how they first learnt. I'd guess these people are in the minority. The majority will want to do things in the best way, as long as it's not too much effort to learn. I started with KDE and Konqueror because it looked like Windows and I could use it. Now I use GNOME 2 and the command line. I found I preferred them, even though they were less like Windows.
What I'm trying to say is, I'd guess most people have a "natural level" of sorts, the place between Enlightenment/FluxBox and KDE where they settle naturally. We have to cater for everybody and we do - but that's not an argument against things that aren't like Windows.
I would love to hear from you or someone else what kind of a replacement you can offer for a user-friendly file manager.
Should be obvious but the command line. For people who can still learn new things (ie most Linux users) the CLI is usually just as good a replacement, hence the fact that we all use it.
Being an experienced UNIX user, I do most things from the console, but many times I find myself using konqueror just because it is more efficient for the specific task. For example, a recursive copy of a local directory tree to an ftp server where I also want to rename stuff on the fly.
Wow, that's a pretty advanced example. I'd guess you could do that using the command line too, but if Konqueror is easier for you then more power to you. I personally find the cli easier for everything, but then I know bash scripting. The only feature I wish it had was proper undelete (I must look into that old slashdot story about this).
I guess my point is very simple: regular users (and this means 99% of the potential users) need a powerfull yet user-friendly file manager
Yes, and it's a good point, but you missed a bit - they need a GUI file manager at the start, maybe they stick with it, maybe they don't.
On the GNOME2 prefs thing, yeah yeah, it's a hot issue. Personally, I found GNOME1.4 to be a hideous mess, and love the clean feel of GNOME2. But when somebody is pleased with something, they don't go around flaming people do they, so they don't get any attention (i was a convert from kde3)
Some peoples issues, in case this whole trollfest passed you by:
The lack of preferences: GNOME2 had a lot of stuff removed. Most of it was pointless bloat, BNC binary clock anybody? Some of it were features that were valued by their users, but were so called "crack" features as far as the gnome2 developers were concerned, ie they existed purely to satisfy a tiny minority of people. The theory went that so called "crack" options (a good example would be the, please break my clipboard again pref in KDE3) were ususally just to either work around bugs, or to make up for the fact that some people had got used to a behaviour that actually made no sense. Every pref has a cost in terms of UI bloat, so they were removed.
Some such prefs will get back in to gnome2. If people can make a convincing case for bringing them back (and "well I liked it" is not classed as convincing) then they could well be brought back. But they don't want the fast and clean v2 to regress to the bag of bloat that was 1.4
GConf. I dunno why people poke this so much. For those who don't know, it's kind of like a registry. Unfortunately the word "registry" is a loaded term, because only Windows has ever had one, and the Windows registry really sucks. GConf is not like that. For starters, the keys are all documented, and they are all stored in text files in your home directory (i believe xml by default). It's well organized (mostly). No, it doesn't need a daemon, it's just most apps use it because that means the configuration and the app can be logically separate - ie you can reconfigure an app while running not just from the config panel, but also from the command line, the GConf editor, a remote machine etc. I think GConf is a great idea, and I wish more apps used it, but it is misunderstood a lot. Another reason that it's used is so that you can have "power user prefs" without bloating the UI, the theory being that power users can use the GConf editor. It works quite well really.
Metacity: unfortunately even I (and I generally think the gnome people have the right idea) think Havoc goes too far. Metacity is very, very "thin" indeed. Although it's not true he doesn't implement any new features, the problem is only stuff that's basically very useful to everybody gets in. Other stuff, stuff that's useful only to perhaps some people (like people who find minimize animations irritating) are ignored. Havoc says "if you want to switch off the animation, there's probably something wrong with the animation", and he's right, there is something wrong which is that it's ugly and slow. But some people on the bug commented that "no matter how fast it is, I'd still find it irritating", but Havoc won't even accept patches others have written to add a GConf key! Don't get me wrong, I'm a big fan of Havocs work esp wrt to freedesktop.org standards, but Metacity does need to grow a bit - the far superior window management of Linux to pretty much anything else is actually a selling point I've found when talking to Windows/Mac users.
Attitude of the developers: the GNOME2 developers have (un)fortunately decided that they are not writing a desktop for geeks, Rasterman can do that, they are writing a desktop for non-geeks. As such, they sometimes come out with comments like "normal users would never need that feature, so it's just bloat" (I'm paraphrasing). As you can imagine, most of gnomes users believe that they as real users are more important than some imaginary, potential users in the future, and big flamage results. I'm not going to comment either way, as it's true that a big problem with Linux usability is the "by geeks, for geeks" mentality, but it's also true that projects that don't listen to their users..... are what? Are pointless? Will die? I don't know.
As for Nautilus - well, I'd rather they dropped it and used ROX which has the advantage of not being originally written by idiots, very fast and doesn't kowtow to Explorer. It's the sort of thing you could embed. For many of the current gnome users though, they (like me) just use the command line - the real Linux answer to Explorer. Stuff like Konqui and Nautilus are perhaps best thought of as training wheels.
Well, look on the bright side. If Blunkett is trying so hard to pass this law, it can only mean that Echelon is not as effective as some people thought. I've actually been interviewed by some people who work for the UK govt and showed me some software they'd written that trawled USENET attempting to corrolate posts together and search for patterns. It was quite advanced too (written in python!).
Clearly though the idea that Echelon can hoover up phone/emails and record/scan them is just so much hooey, as I always thought it was. Reassuring in a way.
Hear hear, quite right. A few corrections however:
optimising the protocol for low-bandwidth transports (see the LBX extension)
LBX is basically dead, it didn't work very well when it was alive, and even with it X is still almost unusable over a slow connection:( There's a postmortem on the XFree site for those who are interested.
eliminating the network socket overhead entirely (see the MIT-SHM extension)
As far as I know, the SHM extension is only used for pixmaps, other stuff like events still travel over a socket. Note however that this is a UNIX domain socket, so is basically a direct line to another process via the kernel. It's far more efficient than network sockets, I think people often get confused about this.
transparency and alpha blending, including font antialiasing (see the RENDER extension)
XRender allows for hardware accelerated alpha-blending it is true, but I think the problem is that there isn't any way of getting a pixmap of what is underneath a window, making it not enough for semi-transparent windows. Hopefully soon....
Re:OK, that's it, I'm 'switching'
on
Linux 3.0
·
· Score: 3, Informative
I'm going to buy a new system in a month or so, and a stable linux 3.0 (I've heard the odd/even thing...3.0 is divisible to 1.5...right?) seems like a good excuse to finally install linux.
Welllll, if it's 3.0 that you want to try, I'd wait a few months (at least). For starters, most of the major commercial distros (and for a beginner, you want a commercial distro) have just had major revs and so won't be upgrading their kernels for a while. Linux 3 will a lot of testing before all the wrinkles are ironed out, even after release, so when you start seeing companies like RedHat, Mandrake and SuSE (the one I use) shipping with a 2.6/3.0 based kernel, then you know it'll be ready.
On the other hand, if this story just piqued your interest, and you want to try anyway (2.4.18 is pretty good) then head over to linux.com where they have a good newbie article and a ton of links to help you out. A good boxed commercial distro will come with printed manuals to get you started, and if you get stuck, want to know a command etc (and you will) come say hi to us all on irc://irc.freenode.net/#linuxhelp
However, I'm not clear what feature of Reiser4 it is that you're demonstrating. Could you provide an explanation to accompany your example?
The feature he's referring to is the ability to treat files also as directories, and it promises to be a majorly cool enhancement to Linux that rewrites one of the most basic assumptions of OS design. The idea is that by improving the power of the filing system layer, and boosting the performance of the FS for small files, the need to have databases layered on top (and even/etc/passwd is a database) is eliminated. This in turn leads to a more powerful OS as that power is made generic, so being available to everything. It's better explained by Reiser himself.
One problem - 2.6/3.0 won't have that ability. What it will have is a special system call reiser4() that Hans can play with. You won't be able to "cat/etc/passwd/mike/group" anytime soon, unfortunately, this kind of major change takes a long time to work its way though the system. The reiser4 call will allow Hans to experiment with the new semantics before we even start to think about merging with the actual kernel.
Why is small file performance so important (this is the area where RFS kicks the ass of, well, pretty much everything else)? Because there are quite a lot of files out there which would actually be better stored as lots of small files./etc/passwd is one good example, there are others. The reason they aren't currently stored as files is because traditionally filing systems have sucked when you have lots and lots of very small files, and we're talking like perhaps 5 byte files here. Reiser4 has some extremely clever algorithms in it, which mean it's good at small files but also large files too.
Of course, this is just the start of a much bigger picture, that'll see the filing system become something akin to a searchable knowledge store. Unfortunately, it's not going to happen quickly. For starters, if you were to suggest to the maintainer of app foo that they should store their data as lots of small files, they's say "no way, some of my users are on ext3, or xfs, or jfs" etc. Reiser has great vision, but he's not the only player in this field, and I have a nasty suspicion that the goal of exploding out large files into filing system structures could prove to be difficult while other filing systems are prevalent. Let's hope not, eh?
One thing I've considered lately might be the possibility of juggling things about a bit - this would not involve altering X at all, so we get to keep our existing investment in the X infrastructure.
What happens is that instead of the application painting the the screen when X requests it, the app uploads a description of the GUI it wants using some kind of XML schema, perhaps a modified XUL, to an ui server (a la display postscript). That XML is then transformed into SVG and rendered using the spiffy 2D acceleration primitives that the XFree team are working on (Xr and Xc or something???). This ui server manages the windows for an app, and if the app requires direct painting they they use XEmbed to make that happen.
This has a number of advantages, namely that being purely vector based it's resolution independant, when an application is working hard the GUI doesn't slow down, by boosting the priority of the ui server you can make the GUI feel much more responsive under load, if an app freezes the GUI does get "damaged" and you could achieve more bandwidth-friendly network transparency.
All this could be done without any modifications to X at all. You'd need to develop some kind of remote DOM synchro technology first, one that could marshal XML Events, but once that was done the rest would be fairly easy.
I don't mean to diss their work, but considering they've spent a year "designing the architecture", where are the docs explaining exactly how it's going to work?
So far the only info on the site are a rundown of the technologies they've "evaluated". However, they talk about using Jabber as a P2P transport - but Jabber is server based. I've not seen any demos of a p2p version of jabber either. Have they actually thought this through?
Is that really how it works in Britain? That's a bit odd when you think how the British tv system worked for so long; No commericals, citizens pay an annual fee. Commercialization of schools seems much more insidious. But with many budgets being tapped these days, where else are they going ot get the money?
Thankfully, no, not yet. It's been talked about a lot though, and a few experiments have been done. I believe one or two "failing" state schools have been taken over by a company, and that sort of thing has indeed gone on. The schools got new buildings (schwag), but had all sorts of conditions placed upon them.
Note that New Labour isn't really left wing at all, they have privatised pretty much everything they possibly could. We still pay for TV because the BBC is far too big and popular to change now.
That's what I'm worried about in a way - these companies seem to be using schools as a kind of playing field for their products. If their products didn't involve any other conditions I'd be fine - no problems with giving out free textbooks etc as long as no favours were expected later. The main problem I see is that a proprietary platform by definition has conditions attached, namely that in future you will be under pressure to upgrade when discounts may not be available, and that it's hard to stop using them etc etc.
No, the logic was correct. You're just an idiot. It was 'worth it' for them to upgrade rather than to completely move their operation over to another OS that didn't have these licensing problems.
That's exactly my point. Whether an OS upgrade is "worth it" often has little to do with the OS itself, and is more to do with how painful would it be to get left behind, how easy would it be to switch to another OS etc. MacOS is not really any different from Windows in terms of licensing, hence all the bitching and moaning about the price of 10.2 - those people clearly didn't think that remaining on 10.1 for ever was really an option.
Yes, Apple may have an agenda but the bottom line is schools are getting free shwag. It's always good to see and I wish more companies would follow suit both in and outside the tech industry.
So you're for the type of school privatization that we're seeing in the UK then, where companies pay for textbooks, buildings, teaching materials in return for students using only branded stuff, and only being able to buy new stationary etc from one location?
I'd think it's pretty easy for local authorities to turn a blind eye to corporates giving schools free stuff, only to later find that their budgets are now dependant upon corporate aid.
Although I would be extremely frustrated in your situation, I don't think that Apple has "ripped you off again." Presumably you thought the upgrade was worth it, otherwise you wouldn't have bought it.
False logic - lots of business have just upgraded to Windows XP due to some fillip of the new licensing regime rather than because they thought it was worth it for instance. If you decided that 10.1 did everything you would ever need and didn't want to upgrade, as time passed your computer would become more and more obsolete, incapable of running newer apps that took advantage of new features added into later versions.
How many people still run Windows 95, even though 98 had few compelling reasons to upgrade (except perhaps real usb support)? The answer - not many. Even Microsoft doesn't support 95 anymore.
Are they just giving away the upgrade or the entire package? If it's the whole OS that's a pretty good deal... heck I'm betting it will sway a lot of people who are thinking of buying a new PC into buying a Mac.
How does that work? If they give you the OS for free, can you go to a shop and buy a cheaper Mac? I was under the impression that the cost of OS X is included and not optional. The only people who would benefit from this are those who already have a Mac with OS X.
How can you compare this with Microsoft bundling Internet Explorer with Windows?
Apple isn't breaking the law.......
Are they not? This isn't a troll, IANAL etc, but isn't giving away a product for below the price you'd normally charge (except in special circumstances) market dumping, and as such illegal?
From what I recall, it wasn't the integration of IE with Windows that landed Microsoft in court originally, it was their practice of giving it away for free in order to get an advantage over Netscape.
In most parts of the world I'm pretty sure market dumping is illegal. The catch could be what the "cost price" of software is - I'd guess the market price they are charging for it, but it could be argued software costs nothing to produce so can legally be given away for free.
While I agree with your post the same could be said of linux, linux very rarely 'innovates'
I'd have a few issues with this statement:
For starters, what is "linux"? People say things like, Organization X doesn't innovate, but that's nonsensical. Individuals innovate, at least in my definition of innovation. A company is just a collection of individuals. "Linux" the movement cannot innovate, but the people inside it can.
Secondly, what is innovation? There are some pretty clever optimizations in the Linux kernel, are they innovative? Apt-get is nothing remarkable technology wise, yet it's a memorable feature that is often one of the most loved, and nothing else has it. Is apt innovative? What about the movement itself? Is the GPL innovative?
OK, so if people who work on Linux can innovate, why don't they? Well arguably they do to some extent, there are lots of other replies to your post that list innovations. However, I think you'll find the biggest reason is because we want to build something that works, and works well, not a research OS. Right now Linux is still catching up on the desktop side of things (which is where you see the most stuff), so the priority is on building a good desktop OS, not on coming up with fantastic "innovations" that may or may not actually be any good.
On the server side, where Linux is somewhat stronger and more advanced, you do indeed see "innovations", although we may have differing definitions. And finally, you can see innovations in the work of people like Hans Reiser, who are pushing for OS design for the Linux community.
Try not to make sweeping generalizations. As Linux catches up with the competition, you'll start seeing more innovation, simply because that's where we'd have to go next.
I believe you can get special boxes that jam mobile phone signals, preventing them from going off (or making calls). Some high class restaurants use them in London I think, it amazes me they aren't practically mandantory for cinemas these days.
Slashdot Mirror
Network and file format compatability are lemons, especially considering the open source crew have done 90% of the work for them. Software compatability is what matters, being able to run custom apps, and of course all that other software that exists out there on the net that uses the Win32 GUI.
A very valid point. I blame the morons who didn't know how to write portable code, as well as the morons who didn't require it in the first place. Yes, the Win32 GUI isn't going to port, but if the app was well written, porting it would be a viable option.
Have you ever tried to write portable software of any complexity at all? If you think it's merely a case of saying "oh, well, I'll make this software totally cross platform" then you need to get more experience of programming in the real world. Writing cross platform software is hard with a capital H. Notice how much duplication there is in most portable projects? Mozilla and OpenOffice both implement their own widget toolkits, portability libraries, object models etc. The GIMP relies upon a dodgy windows port of GTK. The fact is, that writing cross platform software that is high quality on all platforms is a nightmare, and for custom business apps, where ALL your computers are Windows, needlessly spending money on portability would make you uncompetitive. So all your competitors are using software X to help them get ahead, but you're not going to because it won't run on an OS you don't use? Sounds like a going out of business strategy to me. The only solution is something like Wine.
So what if the Mac has no equivalent, as long as it can connect to the same CrossOver Office server?
So your solution to running Windows software on the Mac is to load it up with an X server and use a Linux box as well? That could work for Office. That assumes a lot though, like all software being multi-instance. Most of it isn't, because doing all the locking etc of data files needed is also hard and for desktop apps on Windows, historically single user, 99% of the time not needed. Office can do it with a couple of hacks, because they needed it for Citrix originally afaik. Most can't, you need to run a copy of each desktop to get good performance, if it runs at all.
They will drop Windows when there is an alternative that is 200% as good as Windows, 100% compatible, and only 10% of the cost. Oh yeah, it must have excellent online training too.
Linux hasn't reached all those figures yet (though it's well on the way), and has no online training of any merit (or indeed online help all too often). When it has all those, then companies will switch.
A medium sized company can easily have 5000 desktops. The average price of a Mac is I'd guess about £1000, so that's a cool £5,000,000 (about $7,500,000) just to replace hardware that already worked? You might as well pay the fees to Microsoft, that's almost certainly cheaper. And don't forget that most businesses have at least 1 or 2 custom apps.
The obvious solution is Linux - with a decent set of administators Linux is within a year of being just great on the corporate desktop. The final usability problems are being hammered at a truly astonishing rate, and with tricks like CrossOver Office Server you can pay for 1 copy of Office (I guess it'd work with other programs too) but have it serve hundreds of desktops. Wine is so critical in these areas, for custom business apps, and the Mac has no equivalent, probably won't for some time, if ever.
here she is
I can't see what the fuss is about myself....
Read the whitepaper on namesys.com - he'll be introduced set theoretic naming later on, allowing for associative databases (not relational like longhorn).
Yes, I think Explorer can do something similar. It's not quite the same, but as I'm not a heavy Finder user I'll have to take your word for it....
The Finder doesn't take up memory with caching all the directory structures, just the most recent. So the rendering is slow due to the Finder reading the icons + positions + the directory listing and all the other info from the disk.
Hmm? Neither does Rox, but it's still very fast. I don't know how Tom Leonard does it, but I recall seeing a brief explanation. It's got some pretty clever internals to make it extremely fast - which is why people associate rox with fast. "Buy a faster disk" isn't really a good solution compared to "make the finder faster".
Ahh, now we're getting somewhere. This has been a problem since OS X, because type and creator codes are no longer required, and thus files dont' always have them. The most common types such as .png, .jpg, etc. are supposed to be noticed by the finder and passed to QuickTime for this same sort of detection, all behind the scenes. [snip]
Hmm, sounds like it was good, albiet very complex. 10/10 for flexibility. Wouldn't running apps themselves have quite a bit of overhead though if you wanted to compile say a view with metadata in it? file is very fast as it just uses a combination of lookup tables with binary regexs iirc.
A simple RTFM solves this. Use the technique I described above, type the name until the highlight matches, and Apple-O, or any other keyboard shortcut you'd like, depending of course, on what you want to do with the file. You can actually move quite fast once you get the hang of it, it comes as natural as typing and using the shift key.
What happens if I have two files called "Business Plan" and "Business Diary" (let's pretend). Can I type "bu - tab - d - enter"? Or do I have to type "Business D - enter"? This isn't trying to trip you up or anything, I'm just curious.
Ahh, yes this feature coudl stand out a little more but the Finder is no Web Browser and therefore we keep this tucked away under the Go menu. Choose "Connect To Server..." or, for speed-shortcutting, use Apple-K (Konnect, C interferes with Copy)
OK, cool. Is that a VFS type system? ie in Konqueror I can go to audiocd:/ and copy my audio tracks named from a CDDB server as Oggs to my hard disk, and the files will be ripped and compressed transparently. There's nothing magic about that, what's neat is that it looks just like another FS to me. You could argue this is bloat, but I see no reason why it should be slow technically....
You clearly haven't used the Finder much, IamTheRealMike, but then you also seem to not care because it is Apple. This is a theme that came across in your post. Dont' assume you need to tell someone off because they defended their view - if I was to bad mouth the Linux kernel I'd be killed on this website, by all the slashdotters doing the same. Give it another try, look around carefully, Apple does things differently for a reason. You won't be a power user in a day. That's why Apple's interface is great, it does things logically but not the way other interfaces too. There is too much similarity with other file borwsers that limits them in some respects.
Well, good points. I wrote that last paragraph mainly because toupsie is a well known Mac troll, pretty much every post I see from him/her that gets modded up is estolling the Mac and trashing everything else. I'm not a heavy Finder user, I tend to use it on other peoples Macs, and I don't actually hate the Finder or anything, but I was trying to counterbalance toupsie who seemed to think that the Finder was perfection (it isn't) and that Explorer/the CLI/Nautilus/whatever is automatically awful, because it isn't Apple (they aren't)
Well, thanks for the informative reply. I wish all replies to my posts were like that......
You're enthusiasm for Apples products is great, but I feel I give an alternative viewpoint. I've used the Finder, Explorer, Nautilus, Konqueror, Rox, and the command line, so I have quite a bit to compare it to.
I feel the Finder sucks bigtime:
The NeXT style columns view (the default) is awful. I found I couldn't get it to display as much info as Explorer could in the same space, I found that copying between two locations meant I had to open 2 finder windows or engage the rather feeble tree widget. It wastes space, the big icon/preview is very pretty, but 90% of the time useless as I already know what the filetype is, it just takes up a big fat wad of space that could have been used for something else.
It's slow. No really, even on 10.2, I could watch as it rerendered the Finder on a complex directory structure. Quartz Extreme me all you like, I didn't try it with that, as the drivers for the card in the machine I was using didn't support it (the owner had upgraded it himself). Rox is fast. The Finder is slow.
Primitive typing: in Rox if I view the properties of a file, the "file" program will scan it and try to figure out what exactly it is. It'll say for instance "Screenshot.png: PNG image data, 1024 x 768, 8-bit/color RGB, non-interlaced" or "ASCII Text, long lines". The Finder just says "Document" for any type that isn't explicitly registered with it, at least rox tries to guess based on some reasonably smart heuristics
Apparently no Rox/Nautilus type-ahead tab complete. This isn't a "hard" feature, once you know it's there anybody can use it, I've seen die hard Windowsers pick it up in less than 10 seconds. If the Finder has it, it didn't make it particularly obvious. Rox has a great implementation, just hit / and use it like you would the CLI, you can see it scan through the directories as you type, and get visual feedback as it matches. Nautilus2 has something similar though not as slick if you press Ctrl-L
No address bar? I feel sure it can mount FTP drives etc as it can do the iDisk, but there's no obvious place to type in any URLs for that.
You clearly like the Finder toupsie, but then you like anything that is Apple, and hate anything that isn't, this is a theme that comes across in most of your posts. File management is very much a personal thing - don't assume your view is the "right" view. Comments like "I can't explain, you must just use it" don't help your arguments by the way.
But it is a good idea? I'd bet 90% of Linux users are ex-windows users, which by definition meant ex-explorer users. Yet most Linux users I know use the command line. Perhaps because once you've learnt it, it's faster and more efficient?
Yes, but I can bet that an application like Nautilus is necessary if you ever want to have a decent market share on the desktop . By decent I mean 10% or more.
I think it's a shame we feel we have to copy Windows in order to have Real People (tm) use it. Remember, there is no geek/real person divide, only shades in between. As Linux gets easier to use, so more people overcome the learning curve and the community of users grows. Some people will never, ever want to do anything in a way that's different to how they first learnt. I'd guess these people are in the minority. The majority will want to do things in the best way, as long as it's not too much effort to learn. I started with KDE and Konqueror because it looked like Windows and I could use it. Now I use GNOME 2 and the command line. I found I preferred them, even though they were less like Windows.
What I'm trying to say is, I'd guess most people have a "natural level" of sorts, the place between Enlightenment/FluxBox and KDE where they settle naturally. We have to cater for everybody and we do - but that's not an argument against things that aren't like Windows.
I would love to hear from you or someone else what kind of a replacement you can offer for a user-friendly file manager.
Should be obvious but the command line. For people who can still learn new things (ie most Linux users) the CLI is usually just as good a replacement, hence the fact that we all use it.
Being an experienced UNIX user, I do most things from the console, but many times I find myself using konqueror just because it is more efficient for the specific task. For example, a recursive copy of a local directory tree to an ftp server where I also want to rename stuff on the fly.
Wow, that's a pretty advanced example. I'd guess you could do that using the command line too, but if Konqueror is easier for you then more power to you. I personally find the cli easier for everything, but then I know bash scripting. The only feature I wish it had was proper undelete (I must look into that old slashdot story about this).
I guess my point is very simple: regular users (and this means 99% of the potential users) need a powerfull yet user-friendly file manager
Yes, and it's a good point, but you missed a bit - they need a GUI file manager at the start, maybe they stick with it, maybe they don't.
On the GNOME2 prefs thing, yeah yeah, it's a hot issue. Personally, I found GNOME1.4 to be a hideous mess, and love the clean feel of GNOME2. But when somebody is pleased with something, they don't go around flaming people do they, so they don't get any attention (i was a convert from kde3)
Some peoples issues, in case this whole trollfest passed you by:
The lack of preferences: GNOME2 had a lot of stuff removed. Most of it was pointless bloat, BNC binary clock anybody? Some of it were features that were valued by their users, but were so called "crack" features as far as the gnome2 developers were concerned, ie they existed purely to satisfy a tiny minority of people. The theory went that so called "crack" options (a good example would be the, please break my clipboard again pref in KDE3) were ususally just to either work around bugs, or to make up for the fact that some people had got used to a behaviour that actually made no sense. Every pref has a cost in terms of UI bloat, so they were removed.
Some such prefs will get back in to gnome2. If people can make a convincing case for bringing them back (and "well I liked it" is not classed as convincing) then they could well be brought back. But they don't want the fast and clean v2 to regress to the bag of bloat that was 1.4
GConf. I dunno why people poke this so much. For those who don't know, it's kind of like a registry. Unfortunately the word "registry" is a loaded term, because only Windows has ever had one, and the Windows registry really sucks. GConf is not like that. For starters, the keys are all documented, and they are all stored in text files in your home directory (i believe xml by default). It's well organized (mostly). No, it doesn't need a daemon, it's just most apps use it because that means the configuration and the app can be logically separate - ie you can reconfigure an app while running not just from the config panel, but also from the command line, the GConf editor, a remote machine etc. I think GConf is a great idea, and I wish more apps used it, but it is misunderstood a lot. Another reason that it's used is so that you can have "power user prefs" without bloating the UI, the theory being that power users can use the GConf editor. It works quite well really.
Metacity: unfortunately even I (and I generally think the gnome people have the right idea) think Havoc goes too far. Metacity is very, very "thin" indeed. Although it's not true he doesn't implement any new features, the problem is only stuff that's basically very useful to everybody gets in. Other stuff, stuff that's useful only to perhaps some people (like people who find minimize animations irritating) are ignored. Havoc says "if you want to switch off the animation, there's probably something wrong with the animation", and he's right, there is something wrong which is that it's ugly and slow. But some people on the bug commented that "no matter how fast it is, I'd still find it irritating", but Havoc won't even accept patches others have written to add a GConf key! Don't get me wrong, I'm a big fan of Havocs work esp wrt to freedesktop.org standards, but Metacity does need to grow a bit - the far superior window management of Linux to pretty much anything else is actually a selling point I've found when talking to Windows/Mac users.
Attitude of the developers: the GNOME2 developers have (un)fortunately decided that they are not writing a desktop for geeks, Rasterman can do that, they are writing a desktop for non-geeks. As such, they sometimes come out with comments like "normal users would never need that feature, so it's just bloat" (I'm paraphrasing). As you can imagine, most of gnomes users believe that they as real users are more important than some imaginary, potential users in the future, and big flamage results. I'm not going to comment either way, as it's true that a big problem with Linux usability is the "by geeks, for geeks" mentality, but it's also true that projects that don't listen to their users ..... are what? Are pointless? Will die? I don't know.
As for Nautilus - well, I'd rather they dropped it and used ROX which has the advantage of not being originally written by idiots, very fast and doesn't kowtow to Explorer. It's the sort of thing you could embed. For many of the current gnome users though, they (like me) just use the command line - the real Linux answer to Explorer. Stuff like Konqui and Nautilus are perhaps best thought of as training wheels.
Clearly though the idea that Echelon can hoover up phone/emails and record/scan them is just so much hooey, as I always thought it was. Reassuring in a way.
optimising the protocol for low-bandwidth transports (see the LBX extension)
LBX is basically dead, it didn't work very well when it was alive, and even with it X is still almost unusable over a slow connection :( There's a postmortem on the XFree site for those who are interested.
eliminating the network socket overhead entirely (see the MIT-SHM extension)
As far as I know, the SHM extension is only used for pixmaps, other stuff like events still travel over a socket. Note however that this is a UNIX domain socket, so is basically a direct line to another process via the kernel. It's far more efficient than network sockets, I think people often get confused about this.
transparency and alpha blending, including font antialiasing (see the RENDER extension)
XRender allows for hardware accelerated alpha-blending it is true, but I think the problem is that there isn't any way of getting a pixmap of what is underneath a window, making it not enough for semi-transparent windows. Hopefully soon....
Welllll, if it's 3.0 that you want to try, I'd wait a few months (at least). For starters, most of the major commercial distros (and for a beginner, you want a commercial distro) have just had major revs and so won't be upgrading their kernels for a while. Linux 3 will a lot of testing before all the wrinkles are ironed out, even after release, so when you start seeing companies like RedHat, Mandrake and SuSE (the one I use) shipping with a 2.6/3.0 based kernel, then you know it'll be ready.
On the other hand, if this story just piqued your interest, and you want to try anyway (2.4.18 is pretty good) then head over to linux.com where they have a good newbie article and a ton of links to help you out. A good boxed commercial distro will come with printed manuals to get you started, and if you get stuck, want to know a command etc (and you will) come say hi to us all on irc://irc.freenode.net/#linuxhelp
The feature he's referring to is the ability to treat files also as directories, and it promises to be a majorly cool enhancement to Linux that rewrites one of the most basic assumptions of OS design. The idea is that by improving the power of the filing system layer, and boosting the performance of the FS for small files, the need to have databases layered on top (and even /etc/passwd is a database) is eliminated. This in turn leads to a more powerful OS as that power is made generic, so being available to everything. It's better explained by Reiser himself.
One problem - 2.6/3.0 won't have that ability. What it will have is a special system call reiser4() that Hans can play with. You won't be able to "cat /etc/passwd/mike/group" anytime soon, unfortunately, this kind of major change takes a long time to work its way though the system. The reiser4 call will allow Hans to experiment with the new semantics before we even start to think about merging with the actual kernel.
Why is small file performance so important (this is the area where RFS kicks the ass of, well, pretty much everything else)? Because there are quite a lot of files out there which would actually be better stored as lots of small files. /etc/passwd is one good example, there are others. The reason they aren't currently stored as files is because traditionally filing systems have sucked when you have lots and lots of very small files, and we're talking like perhaps 5 byte files here. Reiser4 has some extremely clever algorithms in it, which mean it's good at small files but also large files too.
Of course, this is just the start of a much bigger picture, that'll see the filing system become something akin to a searchable knowledge store. Unfortunately, it's not going to happen quickly. For starters, if you were to suggest to the maintainer of app foo that they should store their data as lots of small files, they's say "no way, some of my users are on ext3, or xfs, or jfs" etc. Reiser has great vision, but he's not the only player in this field, and I have a nasty suspicion that the goal of exploding out large files into filing system structures could prove to be difficult while other filing systems are prevalent. Let's hope not, eh?
What happens is that instead of the application painting the the screen when X requests it, the app uploads a description of the GUI it wants using some kind of XML schema, perhaps a modified XUL, to an ui server (a la display postscript). That XML is then transformed into SVG and rendered using the spiffy 2D acceleration primitives that the XFree team are working on (Xr and Xc or something???). This ui server manages the windows for an app, and if the app requires direct painting they they use XEmbed to make that happen.
This has a number of advantages, namely that being purely vector based it's resolution independant, when an application is working hard the GUI doesn't slow down, by boosting the priority of the ui server you can make the GUI feel much more responsive under load, if an app freezes the GUI does get "damaged" and you could achieve more bandwidth-friendly network transparency.
All this could be done without any modifications to X at all. You'd need to develop some kind of remote DOM synchro technology first, one that could marshal XML Events, but once that was done the rest would be fairly easy.
What do people think about that?
So far the only info on the site are a rundown of the technologies they've "evaluated". However, they talk about using Jabber as a P2P transport - but Jabber is server based. I've not seen any demos of a p2p version of jabber either. Have they actually thought this through?
Thankfully, no, not yet. It's been talked about a lot though, and a few experiments have been done. I believe one or two "failing" state schools have been taken over by a company, and that sort of thing has indeed gone on. The schools got new buildings (schwag), but had all sorts of conditions placed upon them.
Note that New Labour isn't really left wing at all, they have privatised pretty much everything they possibly could. We still pay for TV because the BBC is far too big and popular to change now.
That's what I'm worried about in a way - these companies seem to be using schools as a kind of playing field for their products. If their products didn't involve any other conditions I'd be fine - no problems with giving out free textbooks etc as long as no favours were expected later. The main problem I see is that a proprietary platform by definition has conditions attached, namely that in future you will be under pressure to upgrade when discounts may not be available, and that it's hard to stop using them etc etc.
That's exactly my point. Whether an OS upgrade is "worth it" often has little to do with the OS itself, and is more to do with how painful would it be to get left behind, how easy would it be to switch to another OS etc. MacOS is not really any different from Windows in terms of licensing, hence all the bitching and moaning about the price of 10.2 - those people clearly didn't think that remaining on 10.1 for ever was really an option.
So you're for the type of school privatization that we're seeing in the UK then, where companies pay for textbooks, buildings, teaching materials in return for students using only branded stuff, and only being able to buy new stationary etc from one location?
I'd think it's pretty easy for local authorities to turn a blind eye to corporates giving schools free stuff, only to later find that their budgets are now dependant upon corporate aid.
False logic - lots of business have just upgraded to Windows XP due to some fillip of the new licensing regime rather than because they thought it was worth it for instance. If you decided that 10.1 did everything you would ever need and didn't want to upgrade, as time passed your computer would become more and more obsolete, incapable of running newer apps that took advantage of new features added into later versions.
How many people still run Windows 95, even though 98 had few compelling reasons to upgrade (except perhaps real usb support)? The answer - not many. Even Microsoft doesn't support 95 anymore.
How does that work? If they give you the OS for free, can you go to a shop and buy a cheaper Mac? I was under the impression that the cost of OS X is included and not optional. The only people who would benefit from this are those who already have a Mac with OS X.
Apple isn't breaking the law.......
Are they not? This isn't a troll, IANAL etc, but isn't giving away a product for below the price you'd normally charge (except in special circumstances) market dumping, and as such illegal?
From what I recall, it wasn't the integration of IE with Windows that landed Microsoft in court originally, it was their practice of giving it away for free in order to get an advantage over Netscape.
In most parts of the world I'm pretty sure market dumping is illegal. The catch could be what the "cost price" of software is - I'd guess the market price they are charging for it, but it could be argued software costs nothing to produce so can legally be given away for free.
"Because this program is running as root, the error message below cannot be properly formatted and may appear incorrectly:
Failure while attempting %s"
We may make fun of Microsoft, but that really, really cheesed me off.
I'd have a few issues with this statement:
For starters, what is "linux"? People say things like, Organization X doesn't innovate, but that's nonsensical. Individuals innovate, at least in my definition of innovation. A company is just a collection of individuals. "Linux" the movement cannot innovate, but the people inside it can.
Secondly, what is innovation? There are some pretty clever optimizations in the Linux kernel, are they innovative? Apt-get is nothing remarkable technology wise, yet it's a memorable feature that is often one of the most loved, and nothing else has it. Is apt innovative? What about the movement itself? Is the GPL innovative?
OK, so if people who work on Linux can innovate, why don't they? Well arguably they do to some extent, there are lots of other replies to your post that list innovations. However, I think you'll find the biggest reason is because we want to build something that works, and works well, not a research OS. Right now Linux is still catching up on the desktop side of things (which is where you see the most stuff), so the priority is on building a good desktop OS, not on coming up with fantastic "innovations" that may or may not actually be any good.
On the server side, where Linux is somewhat stronger and more advanced, you do indeed see "innovations", although we may have differing definitions. And finally, you can see innovations in the work of people like Hans Reiser, who are pushing for OS design for the Linux community.
Try not to make sweeping generalizations. As Linux catches up with the competition, you'll start seeing more innovation, simply because that's where we'd have to go next.
Teaching English is always good....... ;)
I believe you can get special boxes that jam mobile phone signals, preventing them from going off (or making calls). Some high class restaurants use them in London I think, it amazes me they aren't practically mandantory for cinemas these days.