> What I've read most of Slashdot users are suggesting is to set > up mirrors of those stuff to let them download it of local network > - great idea. But add to it that you do not need to make yourself > an admin of those mirrors. Just set up an apply process for a mirror > maintainer and let the students maintain the mirrors themselves > (even give the admin-ones way to use BT to mirror).
Actually all you need to do is give them (the maintainers) an account of one of the servers with proper shell access, bandwith/limits, monitoring and so on - that way they can on their own create communieties like gaming.yourorg.edu, linux.yourorg.edu that host mirrors of hot files and also (if you decide to make it public) serve others.
To be honest that is nothing that I've invented myself - most of schools (that are worth of attention) in my country functions that way (including my one).
What for they need BT protocol? I see most of legitimate uses of BT as downloading Linux (or other freenix) ISOs, commercial games demos and other legitimate big files.
If they (the users) are downloading illegal stuff they should be prohibited to do that.
What I've read most of Slashdot users are suggesting is to set up mirrors of those stuff to let them download it of local network - great idea. But add to it that you do not need to make yourself an admin of those mirrors. Just set up an apply process for a mirror maintainer and let the students maintain the mirrors themselves (even give the admin-ones way to use BT to mirror).
That way they can: - learn how to operate such systems - learn how to practically (as in social siences) operate such systems - get some responsibility and management skills
> At least Steve Jobs gave a price ($299) and launch window > (1st Q 2007, funnily enough when Leopard comes out) for his > "launch" item (iTV).
But you are talking about Leopard and I'am talking about new generation of iPods that Zune likes to be in competition to. The fact is you can buy a new shiny iPod right *now*. It will ship to you in few days and after it is shipped you can start spending your money in new version of ITMS.
Now with Zune you can look at some (3) photos... And... And that is basically all you can do with it.:)
I've meant like announce product which is not real at all just to hold competition sales. Hopefully they can do this with Windows/Office. They cannot success with doing it in other markets. After all when you will go after your xmass shopping the stuff you will see that will be iPods in shiny packages, ITMS coupons and so on. Zune is on prototype photos only.
> That is the stupidest feature I have ever heard of. Did they > do any market research that led them to believe this was > something people actually *wanted*? And that it worked the way > they would want it to work?
As I see it this is not a feature for consumers. This is feature for record companies. It is like advertising records. You can sample some records from friend (or stranger - not specified yet) device. These records will expire in some time leaving you only with their names and a link to buy them.:) So it is like promoting new records, same as in radio broadcast - you hear a record for free and then you want to buy it.
It is stupid but I think this is the design goal/assumption of this feature. Since for sure it is not designed to share music freely.
Of course these are just speculations since it can't be otherwise judging from few photos and brief description of functionality.
So I agree with you - it is a complete failure. I don't see any use of that. If I think some record is great I will go to my friend and say "hey put this record on it is rocking" (or something) - and then we will listen to it together on big hifi.
> So let me get this straight... if I have a Zune... and my > friend has a Zune... I can send them a track. Presuming the > units sync up, and wireless works, and the phase of the moon > is correct, because wireless is still basically tin-can-and-string > at times......and he can listen to it for three days before > it self-destructs?
Yeah it is stupid.:) I also wonder how this wifi stuff will work.
> Lame. Useless. Unless my friend and I both decided to take a > risk and go out and get one of these doohickies each, how would > this situation ever come up? Even if you find another Zune owner, > what are the chances he also shares you musical interests?
Oh! Zune is not about your music intrests.:) It is about selling a lot of pop music to a lot of people and also doing a lot of "sharing" (think like commercials). Zune is really not fucused on customer, it is focused on making money of customers and deals with large record companies that don't rally like musical interest, they like *pop*ular music.
This "sharing" (but again - judging on few sentences of description, not real) like misses content that can be shared freely. Like free licensed music (there is such), fair-use music (like you recorded something off air and want to share it).
>... and neither are mentioned, as far as I can see: > Price & battery-life
Maybe because what you see are three photos of a *prototype* device that is not even in production and nobody knows when it will be?:) So these (what you look at) are more like *goals*, not the final effect. Effect which will constitute of the look (ok we see it), price (no mention), specs like battery life (no mention), ease of use (no mention - but I expect independent review for that, it has ipodish wheel for sure - but how does it work?), service around it (think ITMS for iPods). Etc.
So these are really just few photos that say nothing.
> Bit of a pity that: > "The Zune-to-Zune sharing feature may not be available for all > songs on your device, and works only between Zune devices within > wireless range of each other. This feature allows recipients to > play full-length sample tracks up to three times in three days. > Recipients cannot resend music that they have received via the > sharing feature."... as well...
Yeah. Great feature.:))) Share with crippled music...
Also I am suspicious of this wifi stuff. I know wifi is right now quite complicated to use, and it uses lots of battery power (it was not designed especially for low power consumption). Also I wonder about how those devices will autoconfig to work ad-hoc flawlessly.
Well the *photos* do look nice. But nobody have seen the device in wild so they can be as well 3DStudio renders.:) Typical for MS they announce something like it is real - Zune is unreal as for now. Where can I get one? Where can I read independent review of one? Etc. Etc. Etc.
It does not even has branding on it. Just the look of it is not sufficient for branding - look of iPod is a brand itself, this looks like an iPod clone.
Actually I am used to MS strategies like announcing LongHorn with all WinFS and great stuff that just yet still is not here. So I take this announcement like usual. I will think that this is anything worth looking at when I actually can touch one.
Also I have some concerns, especially with:
1. The interface. Those *three* photos look nice but they do not tell anything about how the device works? Is it easy to use? I've used MS PDA (PocketPC) and it was horrible. I am with Palm now which is somewhat less horrible - so I don't se MS mobile products as quite easy to use or reliable.
2. Those ad-hoc wifi networks - how it is going to work? I've used MS PDA with wifi and it was pain. Actually wifi is quite painfull as for now (especially in MS arrangement). So how they are going to wifify those devices? IMHO wifi is not quite good for establishing such semi-PAN networks - BlueTooth is. But wifi - it is a hog on battery for sure.
So conluding - this is not something real. These are few photos even without branding (forget specs, pricing, performance and so on). So it is nothing actually - just to say "hey wait (like few months... or years) and we will have something like iPod is right now!", "don't buy iPod right now - wait (like months... years) for our mythic device" etc.
Quite unfair but at least in this market we know who rocks with launches that are real launches for real existing products and services...
> For example, my name is "foo" and my number is "42". I know my number > and the state, who assigned that number in the first place, knows > that number. If I give my name as "foo" to Xanga and they ask for > my PESEL, then they will also know my number once I've been confirmed.
But what is the problem with that some site will know your name and PESEL?
> Xanga won't immediately know "foo" is "42" until some state agency, > ultimately, confirms it. But, the instant my identity is confirmed, > Xanga will also know my PESEL. At such time, the PESEL becomes useless > for identification because someone else (lots of someone else's actually) > can now pass the same identity check as if they were me. (It's called 'identity theft'.)
So Xanga is a website that imediately steals your identity? Why you wanted to use it in the first place?
> Neat idea with the UV though. I love the idea > of inkless printing, as long as the paper doesn't > end up being more expensive than gold.
In case of data it gets like quite expensive? How much costs 1GB of data on paper and the whole infrastructure behind it (backup, access time and so on)???
Just get over it, get rid of papers. Digital/electronic devices cope better with data than paper.
> If, as you admit, there is no reasonable way for a website to > enforce minimum-age restrictions,
I live in Poland/Europe. For starters.:)
Here when you are born you get a PESEL number which is date of birth +some ID. The same number is printed on your ID documents whenever you are an adult on a minor.
My point is that only you and the state knows that fe. 198402234214 == Jane Kowalski - so all websites need to do in order to verify age is require that PESEL number and then pass it to another organization that is trusted to send snail mail to the person owning the PESEL number. The company only knows the number (not the data associated to it) the special organization knows the address. Then the organization sends (via snail mail) token to verify in WWW service to the owner of the number (theoretically only the owner is entitled to read his own snail mail).
Of course it would be more expensive than just online registration (by few factors). But it depends on scale - if sending snail mail letter costs you $0.1 and on average you earn $10 on an user and 1/5 registered confirms tokens it is still viable.
That is how our biggest auction site operates (something like eBay) - but they need to verify the real adress and person, not the age. And it somehow works.:)
Easy. The same as with paper documents. Put them into proper envelopes and boxes and into shelves in some offsite magazine. There are loads of established paper documents storage systems - you label it, put it into database, do monthly check and retire old stuff etc.
You don't need to have quick access to these CDs, you have digital copies on servers so you just need it in emergency.
You need normal storage same as for paper documents.
> I applaud any attempt at open-sourcing software but I would > worry about the quality of the code if their primary app > is in this much of a mess.
Maybe they assume that most Windows users are idiots... Real Player for Linux if in fact quite neat application - GNOME style I would say. Real Player for Mac is a bit slow sometimes but again it works and is a little neat application. Only on Windows Real Player is real bloatware changing your settings (associations, putting shit in autostart etc).
On Windows boxes I tend to install Real Alternative which is basically stripped set of Real codecs and browser plugin. But I don't really know if it is legal to use it.
> I've got four pentium 1 class machines running right now, all running linux. > One's a laptop, and I just can't afford a new laptop to replace it right > now, and it does suck for office stuff, although it's fine for coding. > Java's kind of slow on it, but it does the job.
OK - you cannot afford new laptop, but if you could the first thing you've done is replace the old one and use the new. If you do Java for living you wouldn't want your workstation to be slow? It is kind of investment.
> The other three machines are all headless and are in my cabinet. > One runs my DHCP, DNS, and IRC. Another was originally set up for > VOIP, although I don't really use it for anything right now. > The third is my firewall, which runs nothing but the iptables > scripts and raccoon. It does things most cheapo routers won't,
And also the three machines consume power - so if you replace the three with one slick server machine it would return the investment in like a year (since it consumes less power). Also you probably don't mind the fan noise from three old PCs - I do.
> since it has the full flexibility of iptables, and it does come in handy.
Heh, same as my router. It has iptables (2.4-linux-uclibc based), telnet interface and all that stuff.
Yeah. Bassicaly - I'am THE MAINTAINER. Screw you guys using GNOME - I don't like GNOME so I'll drop it. It is too hard to maintain - go maintain it yourself. Now that is what I expect of STABLE distro.:) Like *I* don't like whatever you use - go screw yourself.
Phheeeh.:))) It was also a nice point about Slackware.
Disclaimer: I don't use GNOME directly, I use various parts of it. And I also use KDE - you know, on Linux/X11 you can use both of them - only the apps you like. But not on Slackware.
I've meant the 7.1 release. Why I would use 6.9 that works *worst* with my hardware than 7.1? No mind the Xgl/compiz stuff and so on.:) I know - it is now THE-UNIX-WAY to use decent software that does its work. You *NEED* to use old stuff because if you won't THINGS-WILL-BREAK. Not to mention you haven't specified what is wrong with new versions so you need to use old ones.
> Slackware is still my distro of choice. It's utterly stable, and it just works.
Good for you.
> I actually like the text-based install: it's the right technology for what it > does. What more do you need?
I am not bashing Slackware or whatever. I'll just try to answer your question - what do I need from an installer? Well:
- Remote instalation (when you can controll install process remotely via network), be it VNC, sshd with text mode - whatever. just not to need to sit in front of the box (which is sometimes impossible). Fedora/RedHat installer (Anaconda) has nice feature that is instalation via VNC screen over the network. Great to install Linux remotly at friends machine - I just give them boot image and I install it remotely.
- Automated/unattended installs. Now to be honest - I know shit about Slackware installer. Does it offer unattended install mode (possibly single shell script that just does everything)? Anaconda has *great* features regarding this. You can do virtually anything you whish - automated partitioning? network setup? system presetup? package selection? Yep - all is here just to use.
> Besides, it really will run on anything.
But what kind of point is that? For me it is worthless. On decent workstation I would use decent system. On older one I will use specialized one. Slackware cant fit all. Maybe you mean that it will work on large range of various PCs - OK, but it is not all the stuff Linux can run. And I (only my opinion) find it pointless to run Linux on old PCs. Really - I just throw them out and get decent one that lets me work comfortably. I've used to have old-PC-router, but it sucked - now I have full fledged wireless-adsl-modem-router switch and it does the work better. I've used to have old-PC-server - it died. Now I have normal PC with large disks and decent memory and it finally works as expected.
Linux giving new life to old-PCs is a myth. It is in fact a lot slower at common tasks than f.e. Windows 98 with Office 97 - so why bother?
> Turn on Xgl. Seriously. I run it on every one of my workstations > now, and I've gotten so use to the enhanced interface I feel like > I'll get eye cancer if I use a 'non Xgl'd' system for more than > five minutes.
A bit off-topic.
I have mixed feelings about Xgl (compiz-quinn exactly - Xgl is not something you directly see).
Recently I've been playing with it and it is great - I do not at all mean eye candy - I mean *functionalities* that I get. I see live pictures of windows as I switch them. I can swith windows seeing all of them miniaturized with live previews. I can peek into desktop to the right/left - really nice. And it is not only eyecandy - screw bouncing menus and windows - it really does not give you anything but more bloat to the eye.
But on the other had when I use plain X11 with WindowMaker it is a lot more snappier and it just get its work done. My point is that now Xgl/compiz is fine but it uses rather poor window manager. On my laptop when I use Xgl I can't watch movies and use OpenGL (i915 chip).
On my media-center system in living room I have quite powerfull nvidia card and it copes with Xgl perfectly (movies do work, same OpenGL).
The gallery lacks movies. At least when it comes to current desktops - each (OSX, Linux via Xgl, Vista) has nice 3D effects. For me it is quite a move forward. I don't mean eyecandy - screw that. I use OSX on regular basics and recently Xgl on Linux - on one workstation (my media center pc running Linux) I recently installed XFCE with compiz-quinn instead of ratpoison I and don't look back. Especially window switching with miniaturized windows (scale module) is great.
Ok, I was a little hostile with my words. Sorry.:)
(...)
> It *does* look awfully nice, nicer than most X11 WM > implementations of virtual desktops so far that I've seen.
By look you mean aesthetics or what? And the *features* (I really care for them - not the look).
> Having live previews of your applications (movies that > continue playing, etc) is a great feature, and you can > move them between desktops while they're updating live.
Exactly as I am now doing in Linux with Xgl. Maybe that is why I find the phrase "quantum leap" as a bit extragerating. Have a peak:
> Also, the system will automatically switch you to the relevant desktop > when you click on an app that isn't running on the current one.
You mean clicking on app icon that isn't running on current desktop or what (since you can't click an app that is on different desktop)? Well you can do this on X11 WMs also.
> X can definitely do live previews, *if* you have Composite and a decent > compmgr (like compiz) and something like Xgl or AIGLX.
I have it.:)
> However, these technologies are still in their infancy and far from > ready for mass consumption,
Oh and Leopard is in mass consumption really. Where to buy it?
> and many of the video cards lack the proper support for accelerating > all the nifty 3D goodness that the new toys require.
My quite old nvidia does it perfectly. Geez it is only some 2D effects and very few real 3D.
> As usual, Apple is doing a good job, with some (in hindsight) obvious > improvements.
Which are? You've named live previews (yup we have it) and switching to the desktop the app runs (yup we also have it).
> It'll be fun to see how soon we have the same features > implemented on Linux, in X.
From these two features that you described I actually use them right now.
I think you don't get me - I *do* think that Apple will improve this matter in their way (nice UI, user-friendly and polished) but I can't see it as a "quantum leap". Really I've used these nice new shiny (work)Spaces for like 5 years and I don't see anything revolutionary in Apple implementation.
There are some issues with virtual desktops that Apple could have thought of, like: - window matching (when you prior decide where you want certain app to appear) - notifications from windows that are not present on current desktop - saving sessions - launching default sets of apps on each desktop on session startup
Probably few others. But that approach starts problems with older apps that were not designed with such virtual desktops in mind and may be incompatible - we will see how Apple copes with that (I belive they will manage somehow to produce something new).
Concluding: yes I see some innovation to an old technology, no I don't see it as "quantum improvement". Do you see it as "quantum improvement"?
Oh and what is really revolutionary is the Time Machine technology - I'am really curious about that.
> More likely it's an extension of the versioning system available in Aperture. > It wouldn't be the first time Apple has taken the functionality of an > application and extended it throughout the OS.
It is unlikely for me. Problem with versioning is that the versioning itself is not a problem.:) The algorithm is extremely easy - you just save each file version (with metadata regarding when it was modified, when deleted and probably something elese) and file delete. Very easy to accomplish.
The problem you face with this approach is that you quickly run out of space so the problem is not the versioning mechanism (very simple) but the storage system under it. ZFS is a storage system - it deals with most problems of storage in modern way. It has pooling - a technique that allows to store only small changes - f.e. you have 50MB file, save another version of it in which only 2MB are changed - with pooling you consume only 2MB of space (the changes) - very simplified but it is how it works. Also ZFS supports seamles replicating between physical devices - such like described in the keynote - just plug in external media (f.e. a hard disk) and your backup data gets replicated (archived).
Also apps like Aperture are specialised while extending the metaphor to entire system is not.
So my bets are on ZFS. Especially that they have flirted with Sun on this release and included DTrace which is official.:)
> Given that this is a developers' conference they would > have said ZFS if it were ZFS.
What? The keynote? I don't think so - why he was speaking about some silly Mail.app stationery to developers? Does anyone of them cares about Mail.app stationery?:) They speak just about new exciting features. New exciting feature is "the system will ease the backups" - not "the system will include ZFS".
> The one thing that worries me about Spaces is that the website implies that > you might only be able to have an app running in one window. (Implied by the > fact that you can click on something in the doc and go right to that app's > "space" - I'll admit, I've wanted to do this.) What if I have one Word doc > that goes with this stuff, and one that goes with this other stuff?
I belive it will be done like in X11 (think Linux etc.) window managers. The child window will show on the same workspace as parent window. Think like when you have one Word (the parent) open and then you open another document in new window (the child) it will appear on the same (work)Space - this is how X11 window managers usually deal with it.
> now if file versioning was in Linux natively, or Windows, or OS/2, > or the Amiga, or some other desktop operating system like BeOS I'd > think you'd have a point.
I don't think file versioning is the issue here. Simple file versioning can be done with Linux very easy. For example just LD_PRELOAD an library that will intercept all system calls that unlink or overwrite a file and make it save the file in hiden file like mydoc.foo ->.mydoc.foo.ver[1],.mydoc.foo.ver[2] and do a link to the original version. That is not a problem really.
Problem is the storage under such mechanism - it will consume space very quickly (think of saving 50MB file 100 times). So what is different here is at storage level - I belive they use some algorithms like pooling files or even pooling filesystem sectors (think like Fosil/Venti in Plan9 - google for it - great reading) and so on. So I think they are building such functionality on ZFS filesystem (which I would call storage since it is much more than FS). That are only my assumptions.
Still I have problems to swallow their promises to "backup every file in the system" - it is not quite possible without some quite large storage. Maybe for common user needs it is possible, but thinking of usual Mac strenghts (multimedia, design etc.) - there are some really big files here. It just needs a lot of space to back up every version and copy of f.e. big multi-layered Photoshop files for print.
> In all fairness, Leopard's Spaces implementation looks like > a quantum improvement on other virtual desktop managers > I've used. (Granted, it's been awhile since I tried any since > I was never very satisfied.)
Refering to description on apple.com and the slides from keynote why the fuck you think it is an improvement? They just stated that "hey we have this new thing called Spaces and it is like virtual desktops". They have not said nor shown any details of the implementation itself and it is bit to early to judge on leaked copies reviews. So why the fuck you consider them to be better? Any solid argument? Because now you just *belive* they are. I don't see any difference between virtual desktops in virtually *any* X11 window manager (GNOME, KDE, XFCE whatever) - there are here (on X11) and I use them and that is nothing really sophisticated. For fuck sake it is only like having few desktops and a pager that displays their contents. Really - nothing macish about it. Maybe they will add some nice 3D effects - but the basics are the same.
> None of the other VDMs I recall were quite "Mac-like" > enough--by that I don't mean flashy and animated, > but easy to use and understand.
What it is here to understand? Like few virtual displays so you can switch between them and keep different windows on each? Really...
I use Macs on daily basis (also I use X11 based desktops) and I don't see anything new from these slides - maybe I need to *belive*. But I am far from calling it "quantum leap" - maybe they added some nice eyecandy to it though.
Slashdot Mirror
> What I've read most of Slashdot users are suggesting is to set
> up mirrors of those stuff to let them download it of local network
> - great idea. But add to it that you do not need to make yourself
> an admin of those mirrors. Just set up an apply process for a mirror
> maintainer and let the students maintain the mirrors themselves
> (even give the admin-ones way to use BT to mirror).
Actually all you need to do is give them (the maintainers) an account of one of the servers with proper shell access, bandwith/limits, monitoring and so on - that way they can on their own create communieties like gaming.yourorg.edu, linux.yourorg.edu that host mirrors of hot files and also (if you decide to make it public) serve others.
To be honest that is nothing that I've invented myself - most of schools (that are worth of attention) in my country functions that way (including my one).
What for they need BT protocol? I see most of legitimate uses of BT as downloading Linux (or other freenix) ISOs, commercial games demos and other legitimate big files.
If they (the users) are downloading illegal stuff they should be prohibited to do that.
What I've read most of Slashdot users are suggesting is to set up mirrors of those stuff to let them download it of local network - great idea. But add to it that you do not need to make yourself an admin of those mirrors. Just set up an apply process for a mirror maintainer and let the students maintain the mirrors themselves (even give the admin-ones way to use BT to mirror).
That way they can:
- learn how to operate such systems
- learn how to practically (as in social siences) operate such systems
- get some responsibility and management skills
> At least Steve Jobs gave a price ($299) and launch window
:)
> (1st Q 2007, funnily enough when Leopard comes out) for his
> "launch" item (iTV).
But you are talking about Leopard and I'am talking about new generation of iPods that Zune likes to be in competition to. The fact is you can buy a new shiny iPod right *now*. It will ship to you in few days and after it is shipped you can start spending your money in new version of ITMS.
Now with Zune you can look at some (3) photos... And... And that is basically all you can do with it.
I've meant like announce product which is not real at all just to hold competition sales. Hopefully they can do this with Windows/Office. They cannot success with doing it in other markets. After all when you will go after your xmass shopping the stuff you will see that will be iPods in shiny packages, ITMS coupons and so on. Zune is on prototype photos only.
> That is the stupidest feature I have ever heard of. Did they
:) So it is like promoting new records, same as in radio broadcast - you hear a record for free and then you want to buy it.
...and he can listen to it for three days before
:) I also wonder how this wifi stuff will work.
:) It is about selling a lot of pop music to a lot of people and also doing a lot of "sharing" (think like commercials). Zune is really not fucused on customer, it is focused on making money of customers and deals with large record companies that don't rally like musical interest, they like *pop*ular music.
> do any market research that led them to believe this was
> something people actually *wanted*? And that it worked the way
> they would want it to work?
As I see it this is not a feature for consumers. This is feature for record companies. It is like advertising records. You can sample some records from friend (or stranger - not specified yet) device. These records will expire in some time leaving you only with their names and a link to buy them.
It is stupid but I think this is the design goal/assumption of this feature. Since for sure it is not designed to share music freely.
Of course these are just speculations since it can't be otherwise judging from few photos and brief description of functionality.
So I agree with you - it is a complete failure. I don't see any use of that. If I think some record is great I will go to my friend and say "hey put this record on it is rocking" (or something) - and then we will listen to it together on big hifi.
> So let me get this straight... if I have a Zune... and my
> friend has a Zune... I can send them a track. Presuming the
> units sync up, and wireless works, and the phase of the moon
> is correct, because wireless is still basically tin-can-and-string
> at times...
> it self-destructs?
Yeah it is stupid.
> Lame. Useless. Unless my friend and I both decided to take a
> risk and go out and get one of these doohickies each, how would
> this situation ever come up? Even if you find another Zune owner,
> what are the chances he also shares you musical interests?
Oh! Zune is not about your music intrests.
This "sharing" (but again - judging on few sentences of description, not real) like misses content that can be shared freely. Like free licensed music (there is such), fair-use music (like you recorded something off air and want to share it).
> ... and neither are mentioned, as far as I can see:
:) So these (what you look at) are more like *goals*, not the final effect. Effect which will constitute of the look (ok we see it), price (no mention), specs like battery life (no mention), ease of use (no mention - but I expect independent review for that, it has ipodish wheel for sure - but how does it work?), service around it (think ITMS for iPods). Etc.
... as well...
:))) Share with crippled music...
> Price & battery-life
Maybe because what you see are three photos of a *prototype* device that is not even in production and nobody knows when it will be?
So these are really just few photos that say nothing.
> Bit of a pity that:
> "The Zune-to-Zune sharing feature may not be available for all
> songs on your device, and works only between Zune devices within
> wireless range of each other. This feature allows recipients to
> play full-length sample tracks up to three times in three days.
> Recipients cannot resend music that they have received via the
> sharing feature."
Yeah. Great feature.
Also I am suspicious of this wifi stuff. I know wifi is right now quite complicated to use, and it uses lots of battery power (it was not designed especially for low power consumption). Also I wonder about how those devices will autoconfig to work ad-hoc flawlessly.
Well the *photos* do look nice. But nobody have seen the device in wild so they can be as well 3DStudio renders. :) Typical for MS they announce something like it is real - Zune is unreal as for now. Where can I get one? Where can I read independent review of one? Etc. Etc. Etc.
It does not even has branding on it. Just the look of it is not sufficient for branding - look of iPod is a brand itself, this looks like an iPod clone.
Actually I am used to MS strategies like announcing LongHorn with all WinFS and great stuff that just yet still is not here. So I take this announcement like usual. I will think that this is anything worth looking at when I actually can touch one.
Also I have some concerns, especially with:
1. The interface. Those *three* photos look nice but they do not tell anything about how the device works? Is it easy to use? I've used MS PDA (PocketPC) and it was horrible. I am with Palm now which is somewhat less horrible - so I don't se MS mobile products as quite easy to use or reliable.
2. Those ad-hoc wifi networks - how it is going to work? I've used MS PDA with wifi and it was pain. Actually wifi is quite painfull as for now (especially in MS arrangement). So how they are going to wifify those devices? IMHO wifi is not quite good for establishing such semi-PAN networks - BlueTooth is. But wifi - it is a hog on battery for sure.
So conluding - this is not something real. These are few photos even without branding (forget specs, pricing, performance and so on). So it is nothing actually - just to say "hey wait (like few months... or years) and we will have something like iPod is right now!", "don't buy iPod right now - wait (like months... years) for our mythic device" etc.
Quite unfair but at least in this market we know who rocks with launches that are real launches for real existing products and services...
> For example, my name is "foo" and my number is "42". I know my number
> and the state, who assigned that number in the first place, knows
> that number. If I give my name as "foo" to Xanga and they ask for
> my PESEL, then they will also know my number once I've been confirmed.
But what is the problem with that some site will know your name and PESEL?
> Xanga won't immediately know "foo" is "42" until some state agency,
> ultimately, confirms it. But, the instant my identity is confirmed,
> Xanga will also know my PESEL. At such time, the PESEL becomes useless
> for identification because someone else (lots of someone else's actually)
> can now pass the same identity check as if they were me. (It's called 'identity theft'.)
So Xanga is a website that imediately steals your identity? Why you wanted to use it in the first place?
> Neat idea with the UV though. I love the idea
> of inkless printing, as long as the paper doesn't
> end up being more expensive than gold.
In case of data it gets like quite expensive? How much costs 1GB of data on paper and the whole infrastructure behind it (backup, access time and so on)???
Just get over it, get rid of papers. Digital/electronic devices cope better with data than paper.
> If, as you admit, there is no reasonable way for a website to
:)
:)
> enforce minimum-age restrictions,
I live in Poland/Europe. For starters.
Here when you are born you get a PESEL number which is date of birth +some ID. The same number is printed on your ID documents whenever you are an adult on a minor.
My point is that only you and the state knows that fe. 198402234214 == Jane Kowalski - so all websites need to do in order to verify age is require that PESEL number and then pass it to another organization that is trusted to send snail mail to the person owning the PESEL number. The company only knows the number (not the data associated to it) the special organization knows the address. Then the organization sends (via snail mail) token to verify in WWW service to the owner of the number (theoretically only the owner is entitled to read his own snail mail).
Of course it would be more expensive than just online registration (by few factors). But it depends on scale - if sending snail mail letter costs you $0.1 and on average you earn $10 on an user and 1/5 registered confirms tokens it is still viable.
That is how our biggest auction site operates (something like eBay) - but they need to verify the real adress and person, not the age. And it somehow works.
So I think that there may be reasonable ways.
Easy. The same as with paper documents. Put them into proper envelopes and boxes and into shelves in some offsite magazine. There are loads of established paper documents storage systems - you label it, put it into database, do monthly check and retire old stuff etc.
You don't need to have quick access to these CDs, you have digital copies on servers so you just need it in emergency.
You need normal storage same as for paper documents.
> I applaud any attempt at open-sourcing software but I would
i ve.htm :)
> worry about the quality of the code if their primary app
> is in this much of a mess.
Maybe they assume that most Windows users are idiots... Real Player for Linux if in fact quite neat application - GNOME style I would say. Real Player for Mac is a bit slow sometimes but again it works and is a little neat application. Only on Windows Real Player is real bloatware changing your settings (associations, putting shit in autostart etc).
On Windows boxes I tend to install Real Alternative which is basically stripped set of Real codecs and browser plugin. But I don't really know if it is legal to use it.
Great for those novice users of yours:
http://www.free-codecs.com/download/Real_Alternat
> I've got four pentium 1 class machines running right now, all running linux.
> One's a laptop, and I just can't afford a new laptop to replace it right
> now, and it does suck for office stuff, although it's fine for coding.
> Java's kind of slow on it, but it does the job.
OK - you cannot afford new laptop, but if you could the first thing you've done is replace the old one and use the new. If you do Java for living you wouldn't want your workstation to be slow? It is kind of investment.
> The other three machines are all headless and are in my cabinet.
> One runs my DHCP, DNS, and IRC. Another was originally set up for
> VOIP, although I don't really use it for anything right now.
> The third is my firewall, which runs nothing but the iptables
> scripts and raccoon. It does things most cheapo routers won't,
And also the three machines consume power - so if you replace the three with one slick server machine it would return the investment in like a year (since it consumes less power). Also you probably don't mind the fan noise from three old PCs - I do.
> since it has the full flexibility of iptables, and it does come in handy.
Heh, same as my router. It has iptables (2.4-linux-uclibc based), telnet interface and all that stuff.
Yeah. Bassicaly - I'am THE MAINTAINER. Screw you guys using GNOME - I don't like GNOME so I'll drop it. It is too hard to maintain - go maintain it yourself. Now that is what I expect of STABLE distro. :) Like *I* don't like whatever you use - go screw yourself.
:))) It was also a nice point about Slackware.
Phheeeh.
Disclaimer: I don't use GNOME directly, I use various parts of it. And I also use KDE - you know, on Linux/X11 you can use both of them - only the apps you like. But not on Slackware.
I've meant the 7.1 release. Why I would use 6.9 that works *worst* with my hardware than 7.1? No mind the Xgl/compiz stuff and so on. :) I know - it is now THE-UNIX-WAY to use decent software that does its work. You *NEED* to use old stuff because if you won't THINGS-WILL-BREAK. Not to mention you haven't specified what is wrong with new versions so you need to use old ones.
> Slackware is still my distro of choice. It's utterly stable, and it just works.
Good for you.
> I actually like the text-based install: it's the right technology for what it
> does. What more do you need?
I am not bashing Slackware or whatever. I'll just try to answer your question - what do I need from an installer? Well:
- Remote instalation (when you can controll install process remotely via network), be it VNC, sshd with text mode - whatever. just not to need to sit in front of the box (which is sometimes impossible). Fedora/RedHat installer (Anaconda) has nice feature that is instalation via VNC screen over the network. Great to install Linux remotly at friends machine - I just give them boot image and I install it remotely.
- Automated/unattended installs. Now to be honest - I know shit about Slackware installer. Does it offer unattended install mode (possibly single shell script that just does everything)? Anaconda has *great* features regarding this. You can do virtually anything you whish - automated partitioning? network setup? system presetup? package selection? Yep - all is here just to use.
> Besides, it really will run on anything.
But what kind of point is that? For me it is worthless. On decent workstation I would use decent system. On older one I will use specialized one. Slackware cant fit all. Maybe you mean that it will work on large range of various PCs - OK, but it is not all the stuff Linux can run. And I (only my opinion) find it pointless to run Linux on old PCs. Really - I just throw them out and get decent one that lets me work comfortably. I've used to have old-PC-router, but it sucked - now I have full fledged wireless-adsl-modem-router switch and it does the work better. I've used to have old-PC-server - it died. Now I have normal PC with large disks and decent memory and it finally works as expected.
Linux giving new life to old-PCs is a myth. It is in fact a lot slower at common tasks than f.e. Windows 98 with Office 97 - so why bother?
> Turn on Xgl. Seriously. I run it on every one of my workstations
> now, and I've gotten so use to the enhanced interface I feel like
> I'll get eye cancer if I use a 'non Xgl'd' system for more than
> five minutes.
A bit off-topic.
I have mixed feelings about Xgl (compiz-quinn exactly - Xgl is not something you directly see).
Recently I've been playing with it and it is great - I do not at all mean eye candy - I mean *functionalities* that I get. I see live pictures of windows as I switch them. I can swith windows seeing all of them miniaturized with live previews. I can peek into desktop to the right/left - really nice. And it is not only eyecandy - screw bouncing menus and windows - it really does not give you anything but more bloat to the eye.
But on the other had when I use plain X11 with WindowMaker it is a lot more snappier and it just get its work done. My point is that now Xgl/compiz is fine but it uses rather poor window manager. On my laptop when I use Xgl I can't watch movies and use OpenGL (i915 chip).
On my media-center system in living room I have quite powerfull nvidia card and it copes with Xgl perfectly (movies do work, same OpenGL).
> Xorg 6.9 and Xorg 7.0 are functionally the same.
Despite of few rather nice things and few critical (for few people like laptop users) new drivers.
The gallery lacks movies. At least when it comes to current desktops - each (OSX, Linux via Xgl, Vista) has nice 3D effects. For me it is quite a move forward. I don't mean eyecandy - screw that. I use OSX on regular basics and recently Xgl on Linux - on one workstation (my media center pc running Linux) I recently installed XFCE with compiz-quinn instead of ratpoison I and don't look back. Especially window switching with miniaturized windows (scale module) is great.
> Jeez, who pissed in your cornflakes?
:)
:)
Ok, I was a little hostile with my words. Sorry.
(...)
> It *does* look awfully nice, nicer than most X11 WM
> implementations of virtual desktops so far that I've seen.
By look you mean aesthetics or what? And the *features* (I really care for them - not the look).
> Having live previews of your applications (movies that
> continue playing, etc) is a great feature, and you can
> move them between desktops while they're updating live.
Exactly as I am now doing in Linux with Xgl. Maybe that is why I find the phrase "quantum leap" as a bit extragerating. Have a peak:
http://youtube.com/watch?v=2a__LEQWBbg
> Also, the system will automatically switch you to the relevant desktop
> when you click on an app that isn't running on the current one.
You mean clicking on app icon that isn't running on current desktop or what (since you can't click an app that is on different desktop)? Well you can do this on X11 WMs also.
> X can definitely do live previews, *if* you have Composite and a decent
> compmgr (like compiz) and something like Xgl or AIGLX.
I have it.
> However, these technologies are still in their infancy and far from
> ready for mass consumption,
Oh and Leopard is in mass consumption really. Where to buy it?
> and many of the video cards lack the proper support for accelerating
> all the nifty 3D goodness that the new toys require.
My quite old nvidia does it perfectly. Geez it is only some 2D effects and very few real 3D.
> As usual, Apple is doing a good job, with some (in hindsight) obvious
> improvements.
Which are? You've named live previews (yup we have it) and switching to the desktop the app runs (yup we also have it).
> It'll be fun to see how soon we have the same features
> implemented on Linux, in X.
From these two features that you described I actually use them right now.
I think you don't get me - I *do* think that Apple will improve this matter in their way (nice UI, user-friendly and polished) but I can't see it as a "quantum leap". Really I've used these nice new shiny (work)Spaces for like 5 years and I don't see anything revolutionary in Apple implementation.
There are some issues with virtual desktops that Apple could have thought of, like:
- window matching (when you prior decide where you want certain app to appear)
- notifications from windows that are not present on current desktop
- saving sessions - launching default sets of apps on each desktop on session startup
Probably few others. But that approach starts problems with older apps that were not designed with such virtual desktops in mind and may be incompatible - we will see how Apple copes with that (I belive they will manage somehow to produce something new).
Concluding: yes I see some innovation to an old technology, no I don't see it as "quantum improvement". Do you see it as "quantum improvement"?
Oh and what is really revolutionary is the Time Machine technology - I'am really curious about that.
> More likely it's an extension of the versioning system available in Aperture.
:) The algorithm is extremely easy - you just save each file version (with metadata regarding when it was modified, when deleted and probably something elese) and file delete. Very easy to accomplish.
:)
:) They speak just about new exciting features. New exciting feature is "the system will ease the backups" - not "the system will include ZFS".
> It wouldn't be the first time Apple has taken the functionality of an
> application and extended it throughout the OS.
It is unlikely for me. Problem with versioning is that the versioning itself is not a problem.
The problem you face with this approach is that you quickly run out of space so the problem is not the versioning mechanism (very simple) but the storage system under it. ZFS is a storage system - it deals with most problems of storage in modern way. It has pooling - a technique that allows to store only small changes - f.e. you have 50MB file, save another version of it in which only 2MB are changed - with pooling you consume only 2MB of space (the changes) - very simplified but it is how it works. Also ZFS supports seamles replicating between physical devices - such like described in the keynote - just plug in external media (f.e. a hard disk) and your backup data gets replicated (archived).
Also apps like Aperture are specialised while extending the metaphor to entire system is not.
So my bets are on ZFS. Especially that they have flirted with Sun on this release and included DTrace which is official.
> Given that this is a developers' conference they would
> have said ZFS if it were ZFS.
What? The keynote? I don't think so - why he was speaking about some silly Mail.app stationery to developers? Does anyone of them cares about Mail.app stationery?
> The one thing that worries me about Spaces is that the website implies that
> you might only be able to have an app running in one window. (Implied by the
> fact that you can click on something in the doc and go right to that app's
> "space" - I'll admit, I've wanted to do this.) What if I have one Word doc
> that goes with this stuff, and one that goes with this other stuff?
I belive it will be done like in X11 (think Linux etc.) window managers. The child window will show on the same workspace as parent window. Think like when you have one Word (the parent) open and then you open another document in new window (the child) it will appear on the same (work)Space - this is how X11 window managers usually deal with it.
> now if file versioning was in Linux natively, or Windows, or OS/2,
.mydoc.foo.ver[1], .mydoc.foo.ver[2] and do a link to the original version. That is not a problem really.
> or the Amiga, or some other desktop operating system like BeOS I'd
> think you'd have a point.
I don't think file versioning is the issue here. Simple file versioning can be done with Linux very easy. For example just LD_PRELOAD an library that will intercept all system calls that unlink or overwrite a file and make it save the file in hiden file like mydoc.foo ->
Problem is the storage under such mechanism - it will consume space very quickly (think of saving 50MB file 100 times). So what is different here is at storage level - I belive they use some algorithms like pooling files or even pooling filesystem sectors (think like Fosil/Venti in Plan9 - google for it - great reading) and so on. So I think they are building such functionality on ZFS filesystem (which I would call storage since it is much more than FS). That are only my assumptions.
Still I have problems to swallow their promises to "backup every file in the system" - it is not quite possible without some quite large storage. Maybe for common user needs it is possible, but thinking of usual Mac strenghts (multimedia, design etc.) - there are some really big files here. It just needs a lot of space to back up every version and copy of f.e. big multi-layered Photoshop files for print.
> In all fairness, Leopard's Spaces implementation looks like
> a quantum improvement on other virtual desktop managers
> I've used. (Granted, it's been awhile since I tried any since
> I was never very satisfied.)
Refering to description on apple.com and the slides from keynote why the fuck you think it is an improvement? They just stated that "hey we have this new thing called Spaces and it is like virtual desktops". They have not said nor shown any details of the implementation itself and it is bit to early to judge on leaked copies reviews. So why the fuck you consider them to be better? Any solid argument? Because now you just *belive* they are. I don't see any difference between virtual desktops in virtually *any* X11 window manager (GNOME, KDE, XFCE whatever) - there are here (on X11) and I use them and that is nothing really sophisticated. For fuck sake it is only like having few desktops and a pager that displays their contents. Really - nothing macish about it. Maybe they will add some nice 3D effects - but the basics are the same.
> None of the other VDMs I recall were quite "Mac-like"
> enough--by that I don't mean flashy and animated,
> but easy to use and understand.
What it is here to understand? Like few virtual displays so you can switch between them and keep different windows on each? Really...
I use Macs on daily basis (also I use X11 based desktops) and I don't see anything new from these slides - maybe I need to *belive*. But I am far from calling it "quantum leap" - maybe they added some nice eyecandy to it though.
For backward compatibility I guess...