On Sat, 08 Oct 2011 18:53:14 -0700, Darren New wrote:

> On 10/8/2011 14:28, Jim Henderson wrote:
>> Depends on the filesystem in question.  I think the new upcomer 'btrfs'
>> is supposed to be transactional.
> 
> True. I heard recently that one is coming out for Linux. Now, how many
> programs will actually depend on it? And will it be a half-solution like
> disk snapshots are in Linux? :-)

Not sure what you mean about 'disk snapshots'.  As for who's going to 
depend on it, that remains to be seen, as the development isn't finished 
yet.

>>>> And I've yet to see anything more effective than a binary blob as a
>>>> file.
>>>
>>> I'm curious what this sentence is supposed to mean. Binary blobs are
>>> the lowest common denominator, but almost no files actually store a
>>> binary blob.
>>
>> All files are binary blobs.  Some have restricted character sets, but
>> when it comes down to it, a file is nothing more than a collection of
>> bytes.
> 
> Nope. All files are represented as binary blobs, at least in Linux and
> Windows. Name me three types of files that don't have recognizable
> records in them.

It doesn't really matter if they've got recognizable records in them.  
Those records are still "blobs" that have to be interpreted by a program.

It's all just data.  To say otherwise is just silly.  By definition, that 
blob of data has to mean something to *someone*, unless it's just random 
bytes strung together.

> Lots of mainframes had much more sophisticated file systems. PR1ME had
> SQL tables as their basic file system entity. The fact you've never seen
> an OS that has a better file system doesn't mean they don't exist.

I've seen plenty of better filesystems than what's on a typical PC.

> And, indeed, files in Linux and Windows are *not* represented the way
> they are on disk. Both represent files as arrays of bytes with a length
> accurate to the byte. But there's also ACLs, alternate streams (under
> Windows), directories, etc etc etc. If you want to see a language and OS
> that represents files the way they really are, look at FORTH, which
> represents disks as arrays of blocks, and it's up to you to decide which
> files go with which blocks. Or CP/M, the progenitor of our so-wonderful
> ^Z-is-end-of-file custom for text files.
> 
> Memory is a flat array of bytes too. That doesn't mean a language with
> hashtables built in isn't useful.
> 
> (Sorry. You touched a peeve there. ;-)

I'm good at that. ;)

But I'm not saying that because structure can be assigned to it, it's not 
useful.  I'm just pointing out the axiom that data is data is data is 
data.

>> Well, it's more reliable with users who don't have the education on how
>> to restart the service rather than rebooting the system.
> 
> It also assumes that packages which rely on that updated library declare
> that they do, and that the package tells you how to restart the service.
> If you update something in glibc, does Linux know that the apache
> service will run something different?

Sure, the dev who builds the package has to define the dependencies 
correctly.  That's true on any platform.  On Windows (or an AS/400, or a 
Mac) if you link the wrong version of a shared library, badness can 
happen if the dev is depending on the unique behaviour of a function in a 
specific version of the library.

>> It's a matter of design elegance in my book.  Yes, it doesn't really
>> matter if the system reboots a hundred times during the installation.
>> Well, except that I'm used to dealing with a single reboot on OS
>> installs, so each time the system reboots, I stop what I'm working on
>> because I think it's done, and it turns out it's not.
> 
> You know, part of it is the fact that Windows takes better advantage of
> lots of hardware (in the sense that Windows device drivers written by
> the hardware vendor tend to know more about the hardware). Sometimes
> hardware is designed that you can only detect some bits immediately
> after a reset, so rebooting is required to select the right driver out
> of many.

So rather than stop the installation and reboot, it seems it would be 
better to queue those things together so a single reboot deals with them 
all.

But more to the point, what you're essentially saying is that Windows has 
to reboot because the hardware vendor's poor design means their own 
driver can't determine the device correctly unless it's been freshly 
reset.

My BS meter is going off. ;)

> And, honestly, I don't think Windows (Vista) rebooted more than once
> during the last install I did either.

Win 7 did twice, IIRC.

>> But Windows has never been good at telling the user how long
>> something's going to take
> 
> Very true.  Indeed, it bugs me that'll show a progress bar progressing
> even if you're copying over the network and you just pulled out the
> cable or something.  No, really, it's not progressing. Stop flying pages
> from one folder to the next. You've gotten reliable enough that things
> can break without taking down the system to the point where GIF
> animations stop animating just because the network cable is unplugged.

Of course, Microsoft stole the idea of a progress bar that makes no sense 
from Novell.  Just like BSOD (which they embraced, and then 'enhanced' by 
making it 'blue' instead of 'black'). ;)

>  > (to the point that I guess in Win 8, they're
>> going to stop trying to predict things like how long a multiple file
>> copy is going to take to complete).
> 
> Heh.

That made me laugh.  Another area where Linux was ahead in the game 
(arguably, a minor one, but since we're trading barbs. <g>)

Jim

Post a reply to this message

On Sat, 08 Oct 2011 19:20:24 -0700, Darren New wrote:

> On 10/8/2011 18:03, Jim Henderson wrote:
>> On Sat, 08 Oct 2011 18:00:56 -0700, Darren New wrote:
>>
>>> On 10/8/2011 14:38, Jim Henderson wrote:
>>>> Configuration files don't autogenerate other configuration programs.
>>>
>>> Autoconf? :-)
>>
>> Autoconf isn't really a configuration file.  It's a configuration file
>> parser. ;)
> 
> Hey, it's all binary blobs anyway, right? ;-)

Sure.  I suppose you'll point out that the binary structure of an 
executable that actually does something to manipulate the data in another 
file means that the executable is a configuration file.

My point was data is data is data is data, regardless of structure 
assigned to that data.  If you put a "binary blob" in the Windows 
registry, something knows what the structure of that blob means, 
otherwise it's just random data.

>> Though I find that the state of Linux GUIs is improving.  Certainly has
>> since I started using it.
> 
> Oh, tremendously. But they're still a PITA compared to Windows'
> explorer, methinks.

Depends entirely on what you're used to.

Having recently been asked to do some work that requires Windows, I've 
had my own frustrations with the Windows interface and things that don't 
work as efficiently on Windows as they do in Linux.

The same is true for comparing featuresets between Hyper-V and VMware 
Workstation.  For example, in Hyper-V, if you "pause" a machine, it stays 
allocated in memory rather than suspending.  "Pause" means "suspend 
execution".

In VMware, if you "pause" the VM, VMware grabs the state and commits it 
to disk, freeing up the memory for other VMs.  I find VMware's usage 
makes a lot more sense.

Jim

Post a reply to this message

On Sat, 08 Oct 2011 19:19:27 -0700, Darren New wrote:

> On 10/8/2011 18:05, Jim Henderson wrote:
>> I didn't know I could install Adobe Acrobat, Flash, or other third
>> party software from Windows Update. ;)
> 
> You can, if it passes Microsoft certification requirements. I saw
> instructions somewhere for setting it up with Microsoft.

Not really the point, though.

>> (That's actually what I was talking about - not about driver installs)
> 
> And device drivers aren't third party software? Some of the DRM plug-ins
> are similar as well.

Sure, they are - but then again, Linux tends to include a lot more device 
drivers on the media than Windows does.  But sure, ATI and nVidia 
drivers, for openSUSE, are installed from a repository.

>>>> Getting the software makers to agree might take some work,
>>>
>>> And that is the problem. That, and commercial entities don't really
>>> want their software in Windows Update where people could install it
>>> without paying for it. The model really only works for free software.
>>
>> The software in question I'm talking about is available gratis as it
>> is. Like Flash, Acrobat Reader, Java, etc.
> 
> Sure. But people giving away free software want to be in your face about
> it, not as a silent background install. There's a reason that Adobe
> plasters their icon on your desktop on every update regardless of the
> fact you never start it without a file.

Oddly, that doesn't happen with the Linux versions.  Certainly not on 
systems running GNOME3 (which actually won't let you put icons on the 
desktop).

>>>> but then again, Adobe has Acrobat Reader in most distributions'
>>>> official repositories.
>>>
>>> Probably because Adobe doesn't have to deal with it. They just have to
>>> give permission.
>>
>> So why not something like that for Windows, too?
> 
> Sure. Feel free. Go ahead. :-)  The point is that it costs Microsoft
> real money to provide that service, and people generally don't want to
> use that, because Microsoft controls it too much.  Commercial entities
> don't like that, but they'll put up with it if that's the only way to
> get their free software in front of their target audience.

I don't use Windows enough for it to be useful.

But you're saying that it doesn't cost the folks who run Pacman money to 
host repositories?

Jim

Post a reply to this message

On Sat, 08 Oct 2011 19:12:58 -0700, Darren New wrote:

> On 10/8/2011 18:00, Jim Henderson wrote:
>> Well, Andy wants to compare modern Linux distributions with a 10-year
>> old version of Windows.<shrug>
> 
> Sure. I was just offering a counterpoint. Windows and Linux both come on
> one DVD. (I don't know if both 32 and 64 Windows come on the same DVD.
> I've just seen directory listings.) Linux without apps fits on a CD.
> Windows without apps doesn't.

True.

>> Yes, I certainly do.  Software documentation of most kinds absolutely
>> sucks rocks these days.  It tends to focus on what the software does
>> rather than how to do it or why you'd want to use it the way they
>> designed it.
> 
> Or even in enough completeness you could use it. I remember learning APL
> from the APL interpreter manual that came with the computer. How cool is
> that? I learned UNIX by sitting down and reading thru the tome of man
> pages.

Yep, absolutely. :)

Jim

Post a reply to this message

On Sat, 08 Oct 2011 19:15:12 -0700, Darren New wrote:

> On 10/8/2011 18:08, Jim Henderson wrote:
>> Then we agree there.
> 
> I think we agree on this whole message. Nuff said. :-)

Right. :)

>> Sure.  But there are ways of dealing with that, too.  Most computers
>> have USB ports these days, so a USB flash drive can be used (in fact, I
>> did my upgrade from oS 11.4 to 12.1 beta 1 using a flash drive.)
> 
> Sure. But you're not going to upgrade a proprietary for-fee OS like
> Windows from a flash drive. I'm not saying Linux is bad or anything. I'm
> just pointing out *why* it is different, which sadly many people fail
> (or refuse) to notice.

Windows doesn't really seem to have a mechanism for doing an upgrade from 
a flash drive, but yes, usually people are going to do it online.

Jim

Post a reply to this message

> Of course, Microsoft stole the idea of a progress bar that makes no sense
> from Novell.  Just like BSOD (which they embraced, and then 'enhanced' by
> making it 'blue' instead of 'black'). ;)

It least they don't call it a "guru meditation number" any more...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On 10/9/2011 10:15, Jim Henderson wrote:
> Well, I think you probably would - some of the compiled code would
> already be in object form, and the compiler wouldn't have to compile it
> again.

Nope. This was one giant compile of one source file. (Well, lots of nested 
includes, etc.) I didn't have to log back in, restart the compile, or 
anything like that. Plus, of course, on Linux, you'd have a corrupted output 
file, because Make doesn't check that the compile finished, only that the 
object code has a timestamp later than the source code.

> The thing is that on Linux, if you have a problem and report it, there's
> a far better chance it'll be fixed quickly.

Probably true. Unless like you're a giant corporation or something, yeah.

>> I'm not describing faults. I'm describing "catching up with other more
>> popular systems."
>
> It's good to see Windows catching up with Linux, isn't it?  Some of the
> features in Win8 have been available in Linux for years. ;)

And vice versa.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message

On 10/9/2011 10:31, Jim Henderson wrote:
> On Sat, 08 Oct 2011 19:19:27 -0700, Darren New wrote:
>
>> On 10/8/2011 18:05, Jim Henderson wrote:
>>> I didn't know I could install Adobe Acrobat, Flash, or other third
>>> party software from Windows Update. ;)
>>
>> You can, if it passes Microsoft certification requirements. I saw
>> instructions somewhere for setting it up with Microsoft.
>
> Not really the point, though.

Sure, in practice, Linux repos are the first and best source of code, and in 
practice Windows Update is the last place people go to look for updates of 
non-Microsoft code. But as I said, I think that's more the difference 
between FOSS and commercial software than a difference between Linux and 
Windows per se.

Is that your point?

> But you're saying that it doesn't cost the folks who run Pacman money to
> host repositories?

I'm not sure what your point is. It would cost Microsoft way more to host 
upgrades on Windows Update than it would to host free software on a Linux 
repository, even if the bandwidth and storage requirements were identical.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message

On 10/9/2011 10:28, Jim Henderson wrote:
> My point was data is data is data is data, regardless of structure
> assigned to that data.  If you put a "binary blob" in the Windows
> registry, something knows what the structure of that blob means,
> otherwise it's just random data.

OK. I've lost track of why you'd make such a point, but sure.

>>> Though I find that the state of Linux GUIs is improving.  Certainly has
>>> since I started using it.
>>
>> Oh, tremendously. But they're still a PITA compared to Windows'
>> explorer, methinks.
>
> Depends entirely on what you're used to.

I expect there are some areas where you can definitely say "this feature is 
better implemented than that feature", and a bunch where it's much more 
important what you're used to.

But I think it's safe to say that having three different copy/paste buffers 
and having each program decide which key combination to use to address each 
one is inferior to having one copy/paste buffer consistently implemented, 
for example.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message

On 10/9/2011 10:24, Jim Henderson wrote:
> On Sat, 08 Oct 2011 18:53:14 -0700, Darren New wrote:
>
>> On 10/8/2011 14:28, Jim Henderson wrote:
>>> Depends on the filesystem in question.  I think the new upcomer 'btrfs'
>>> is supposed to be transactional.
>>
>> True. I heard recently that one is coming out for Linux. Now, how many
>> programs will actually depend on it? And will it be a half-solution like
>> disk snapshots are in Linux? :-)
>
> Not sure what you mean about 'disk snapshots'.

The ability to take a consistent snapshot of the disk while it's still 
active, and (for example) make a backup of it, without disturbing people 
trying to write to drives you're backing up.

http://www.howtoforge.com/linux_lvm_snapshots

> It doesn't really matter if they've got recognizable records in them.
> Those records are still "blobs" that have to be interpreted by a program.

Well, sure. OK, I'm missing your point, then.  Files are more than just 
blobs of data, or they wouldn't be useful, as you say. They have a meaning 
for the data, even if that meaning isn't stored within the data. But I'm not 
sure what your point is.

> But I'm not saying that because structure can be assigned to it, it's not
> useful.  I'm just pointing out the axiom that data is data is data is
> data.

Sure, I'll grant that. I'm not sure at this point why you're pointing that out.

> So rather than stop the installation and reboot, it seems it would be
> better to queue those things together so a single reboot deals with them
> all.

I think that's basically what they started doing, yes. Remember how often 
NT3.5 rebooted during install?

> But more to the point, what you're essentially saying is that Windows has
> to reboot because the hardware vendor's poor design means their own
> driver can't determine the device correctly unless it's been freshly
> reset.

Possibly. I've read that about some of the hardware. That's part of the 
whole "unified driver" thing going on.

> Of course, Microsoft stole the idea of a progress bar that makes no sense
> from Novell.  Just like BSOD (which they embraced, and then 'enhanced' by
> making it 'blue' instead of 'black'). ;)

Machines have had BSOD since long before MS-Dos was around. :-)

> That made me laugh.  Another area where Linux was ahead in the game
> (arguably, a minor one, but since we're trading barbs.<g>)

Barbs? Trading barbs would imply I care whether Windows is better than Linux 
or vice versa. :-)  Neither is my program, so I don't really care.

-- 
Darren New, San Diego CA, USA (PST)
   How come I never get only one kudo?

Post a reply to this message