>hard to do, no, impossible to do, when its part of the operating
>system.
>
>besides, logitech isn't in the business of martyring itself.  it is the
>parasite, it doesn't have the weight to tell microsoft whats up.  they
>know they need to give customers the best experience.
>
>and yes, i am saying SBS cost me two hard drives.  when SBS was killed
>in task manager, the issues stopped.  which app was at fault really
>isn't the point.
>
Okay, you can blame SBS if you like, but I don't see what that has to do with 
MySQL.  When SBS uses SQLite, it *could* have similar conflicts with other 
software, outside of the control of SBS.  i.e. if the DB file on disk is 
locked, a virus scanner wouldn't be able to read the content and throw a wobbly.

>i don't know, and i don't worry about it, since i don't use plugins
>generally, but i take erland at his word, and like i said i don't hear
>any big reason which outwieghs the benefits.
>
Or any particular reason why a switch is necessary.

>always the case.  you could have plugins break just by a new version on
>SBS that still uses mysql.  frankly, this is not compelling.
>
True.  It's just more likely.

>while tinysbs is probably the driving reason, there are lots of other
>good reasons, like other low-powered hardware, and eliminating extra
>processes and software conflicts.  to me it makes perfect sense.
>
There's loads of extra processes that get spawned from SBS.  If you look in 
server\Bin\MSWin32-x86-multi-thread, you'll see several binary images for file 
type support.  e.g. lame for transcoding to mp3.  Perhaps all of these should 
be re-written to be embedded inside SBS to eliminate extra processes and 
software conflicts?  I think not!  Firewall/AV software can (and does!) cause 
any of these things from working.

>and i am saying why do you say that?  what exactly do you lose that
>outweighs the benefits of sqlite?
>
And I'm saying what are the clear benefits of SQLite?  There's no clear 
benefits of either; and indeed some people will see advantages/disadvantages 
with either, depending on their OS, size of library, AV software, etc.
It's like saying there's a clear advantage for SBS to only support Linux, as 
that is the only OS that it needs to support for Tiny SBS, and Windows 
auto-file change detection is much harder to implement, so some Windows/MAC OS 
functionality will be removed.  Having support for all of them provides choice 
and flexibility.  The user doesn't have to choose which flavour of OS, because 
it automatically uses the right support libraries.

>unlikely.  this would only happen if they started selling a server box
>to go with the player box, (meaning something better than the touch
>server).  besides, they have such a large installed userbase, i just
>can't see it happening, it would alienate way too many users.
>
The current userbase can continue with current SBS software.  They don't have 
to continue to support existing users and Full SBS indefinitely.  They recently 
announced less support for SB3 (endeavor to make new SBS work with older 
products, but not guaranteed).

>i refuse to even test 7.6 yet.  it sounds nightmarish to me.
It seems much improved since this side of xmas.  I'm always happy to test 
bleeding edge, but until recently I found it unuseable.  Critical issues didn't 
seem to be getting fixed.  It's now harder to run SBS 7.5.x and 7.6 in 
parallel, as it can't share the same database/plugins, so it's an all or 
nothing switch forwards (there's no going back).

>i expect that over time they will get sqlite running smoothly on 7.6
I hope so.  I hope that it doesn't go out with any new bugs and any degradation 
of performance, resilience, reliability, and other non-functional requirement 
issues.
Maybe these points will be better; maybe not.  No-one reliably knows, other 
than there will be differences.

Anyway, I seem to be in the vast minority with concerns over SQLite, so fingers 
crossed it will all be okay in the end.

Phil
_______________________________________________
beta mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/beta

Reply via email to