yellow dog 6

[ssam]

http://www.terrasoftsolutions.com/news/ ... 1-12.shtml mentions YDL 6.

does anyone know when this will be out for macs? will the software be upto date enough to compete with other distros (ubuntu, fedora, gentoo)?

[billb]

http://www.terrasoftsolutions.com/news/2007/2007-11-12.shtml mentions YDL 6.

does anyone know when this will be out for macs? will the software be upto date enough to compete with other distros (ubuntu, fedora, gentoo)?

Compete? I don't know ... Is there some specific feature you're hoping for?

I am assuming we'll see an announcement re: YDL 6 on the announcement mailing list, so you might want to subscribe to that so you'll hear about it as soon as possible.

[pablo]

I don't care about 5, 6 or 12.. As long as it works, and it is stable, I will use it. Would I be using something else on my PS3 if they worked? Possibly, just to give it a shot, but now that YDL works without any major issues on my PS3, I will continue using it. It seems stable enough. If it craps out, I will just wipe and re-install.

[ppietro]

http://www.terrasoftsolutions.com/news/2007/2007-11-12.shtml mentions YDL 6.

does anyone know when this will be out for macs? will the software be upto date enough to compete with other distros (ubuntu, fedora, gentoo)?

(What follows is a lot of opinion, mostly based on facts - make of this what you will. )

The interesting thing about this announcement is the CentOS 5 foundation. It has direct bearing on your "up-to-date" question.

If you're not familiar with the whole Red Hat/Fedora/CentOS situation, here's a quick summary:

Red Hat became one of the most popular Linux distros. It attracted freeware enthusiasts and paying businesses alike.

According to a presentation I attended at LinuxWorld by the Fedora team, after a while, this presented a very sticky situation for Red Hat.

Quoting from memory, they said 90% of the new features in Red Hat were being contributed by the enthusiasts who got it for free. However, the company was being sustained by the paying corporate customers.

For a long time, this worked - until the introduced changes hit a break point. The enthusiasts got upset at Red Hat's reticence to introduce major system changes as fast as they were writing them - and the business users got upset at how often Red Hat was updating. This was a catch-22 for Red Hat.

They solved the issue by splitting Red Hat Linux into two separate distros - Red Hat Enterprise Linux (RHEL) and Fedora. RHEL was designed to be stable, and update only as necessary for paying customers. Fedora was designed to be fluid, free, and incorporate all the changes as soon as possible.

YDL 5.0x is based on Fedora.

(There are obviously other differences between the two distros - but bear with me here..)

So - what's CentOS?

Good question. If you recall, RHEL binaries are available to paying customers only. However - since it's under GPL, all of the source code is freely available. CentOS is an attempt to build RHEL from source, removing or building around non-GPL or trademarked code.

What this means for YDL 6.0 is that you'll basically be getting the equivalent of a RHEL release instead of a Fedora release. YDL 6.0 will include select Fedora 7 components, but I would still expect 6.0 to be a little behind the current Ubuntu, Fedora, Gentoo releases - just as RHEL lags behind Fedora.

This is not necessarily a bad thing!!!

RHEL is, by design, very stable. Considering the PS3's ability to do complicated math functions, and work in clusters, I think this is a good thing - it should help Cell-based Linuxes (Linii?) be taken more seriously for scientific and enterprise use.

It should also help prevent some of the hashing between kernels we've seen on the PS3 with 5.0/5.0.1/5.0.2.

And, the way I understand it, you can still update RHEL to be bleeding edge if you like, via Fedora RPMs.

It, of course, remains to be seen what YDL 6.0 will be like. (Well - at least by me! ) It sounds like it should be great!

Cheers,
Paul

[NeoAmsterdam]

http://www.terrasoftsolutions.com/news/2007/2007-11-12.shtml mentions YDL 6.

does anyone know when this will be out for macs?

Based on the beta-test notice I'd lay down 50/50 odds within 6 months, 80% odds in 9 months.

The interesting thing about this announcement is the CentOS 5 foundation.

Indeed - CentOS's (and all RHEL communities') PowerPC efforts have been meager at best. Having used Scientific Linux (a RHEL derivative for x86) for quite some time now, I can tell you that there are times I wished for a PowerPC equivalent. RHEL is robust, stable, well-rounded, and TSS made a good move shifting to it.

That said, the statement from TSS that portions of YDL 6 will be coming from Fedora is confusing: As I recall Ubuntu and Fedora announced that they'd be abandoning the PowerPC. Since YDL 5 effectively shut KDE out of their installation options (among other things), I had to use Fedora 5's PowerPC yum repositories to install packages. If YDL 6's installation neglects package selection (as it did in YDL 5) and Fedora doesn't offer PowerPC RPMs... well, I don't like the thought of spending a half a week compiling the KDE 4 suite after spending a week recompiling the kernel for my pre-G3 Power Mac.

Linuxes (Linii?)

The pluralizing suffix -ii stems from Latin words ending in -us, such as radius and virus becoming radii and virii. IMO the more appropriate pluralizing suffix to use is the x-to-ices conversion, such as suffix, index, and unix becoming suffices, indices, and unices (accent on the first syllable).

By extension, linux would be linuces, not linii (unless you're Microsoft and consider Linux to be a virus).

[billb]

http://www.terrasoftsolutions.com/produ ... very.shtml

[NeoAmsterdam]

http://www.terrasoftsolutions.com/products/ydl/delivery.shtml

:shock: Uhm... Did I say "months"? I meant "days". Yep, "days" Geez, TSS's fast!

[javon27]

i really enjoyed your post Neo. hopefully you can contribute much to our little community, especially on behalf of all the noobs here like me who have only been introduced to linux because of the PS3. Cheers.

[ssam]

looks good

kernel 2.6.23 means we get the new schedulling goodness. .24 would be nice for powerpc tickless (better powersaving), and mainline broadcom support (i guess the yellowdog kernel has this anyway). but .24 has not been out long at all.

download is going a bit slow, but i should get to have a proper look soon.

[aguila]

It is interesting that many standard dictionaries support the singular to plural transition which has been explained very well here. Unfortunately, they don't explain it as clearly.

Sadly I keep finding all kinds of "luminaries" who use the Latin form who seem unaware that consistency requires that if one is choosing that road, then English itself (especially American English), needs to be restructured into Latin for a complete logical construct - in Latin. In other words, the effort to bend or fit Latin into English is at best unwieldy.

The logical constructs in use at the time when Latin was spoken widely is very different from modern English and different even from the modern Romance languages which share an even closer grammatical structure with Latin than even English.

The point is that an exploration into history, language, and usage is important for meaningful communication. Applications, like Word and others, can be useful and helpful but only a human can employ comprehensive understanding to utilize language well.

[billb]

Word.

[Noons]

What was that!?
no really I'm lost

[aguila]

...

Linuxes (Linii?)

The pluralizing suffix -ii stems from Latin words ending in -us, such as radius and virus becoming radii and virii. IMO the more appropriate pluralizing suffix to use is the x-to-ices conversion, such as suffix, index, and unix becoming suffices, indices, and unices (accent on the first syllable).

By extension, linux would be linuces, not linii (unless you're Microsoft and consider Linux to be a virus).

It is interesting that many standard dictionaries support the singular to plural transition which has been explained very well here. Unfortunately, they don't explain it as clearly.

Sadly I keep finding all kinds of "luminaries" who use the Latin form who seem unaware that consistency requires that if one is choosing that road, then English itself (especially American English), needs to be restructured into Latin for a complete logical construct - in Latin. In other words, the effort to bend or fit Latin into English is at best unwieldy. The logical constructs in use at the time when Latin was spoken widely is very different from modern English and different even from the modern Romance languages which share an even closer grammatical structure with Latin than even English.

The point is that an exploration into history, language, and usage is important for meaningful communication. Applications, like Word and others, can be useful and helpful but only a human can employ comprehensive understanding to utilize language well.

[aguila]

...

Linuxes (Linii?)

The pluralizing suffix -ii stems from Latin words ending in -us, such as radius and virus becoming radii and virii. IMO the more appropriate pluralizing suffix to use is the x-to-ices conversion, such as suffix, index, and unix becoming suffices, indices, and unices (accent on the first syllable).

By extension, linux would be linuces, not linii (unless you're Microsoft and consider Linux to be a virus).

According to Merriam Webster,the plural of virus is viruses. See here: http://www.merriam-webster.com/dictionary/virus

It is interesting that Merriam-Webster has no entry for unix, therefore there is no way to explore it's plural via that reference.
The American Heritage Dictionary has an entry for the word unix, but it is in my opinion, incorrectly described. There is no attempt to present a plural version of unix. Their entry is here: http://www.bartleby.com/61/96/U0089625.html

In an attempt to explore this issue of the Latin plural case a bit more thoroughly here is a link which gets into some detail:
http://alt-usage-english.org/excerpts/fxplural.html

Consistency requires that if one is choosing to use Latin grammatical constructs, then we need to restrict ourselves to the grammatical rules in existence when Latin was commonly used. In other words, the effort to bend or fit ancient Latin grammatical constructs and concepts into modern English may be at best unwieldy and probably unworkable as modern languages refer to modern concepts and technologies which the ancients (for all our admiration of them) did not have any experience with or idea of. The logical constructs in use at the time when Latin was spoken widely is very different from modern English and different even from the modern Romance languages which share an even closer grammatical structure with Latin than even English.

The point is that an exploration into history, language, and usage is important for meaningful communication. Applications, like Word and others, can be useful and helpful but only a human can employ comprehensive understanding to utilize language well.

Modern languages can grow and work through a formal exploration of new grammatical usage of technological and scientific concepts and ideas which works it's way into our lives. In the well meaning effort to grasp the problem of exploring evolving technical word usages ahead of a formal declaration of a recommended grammatical construct based on linguistic scholarship, we all risk entering into the completely different problem of creating a niche concept of ideas and definitions which is not Latin, not English nor identifiable with any commonly known or understood language.

We can avoid this if we insist that scholars in the modern languages explore this issue carefully.