What went wrong?

>>56363322
Oracle killed everything good Sun made that's what.

the dotcom bubble took Sun out with it

don't really know what "went wrong" though, still behaves the same as it always has, just oracle doesn't give a fuck about it anymore and really half asses development

>>56364202
>>56363322
I'd say that they killed themselves by refusing to diversify. They only catered to one market, and when they began loosing ground there, they died.

Nobody gives a fuck about Unix anymore. All Unix divisions have died or are now considered legacy.

>>56364226
where else really could they have diversified to? of all the SysVs Solaris is probably the only diverse one of the lot, with a relatively large application base both for desktops and servers, and it saw plenty of use on both in the pre-Oracle days

>>56364275
not true at all, Solaris, AIX and HP-UX are all very much alive, they're just niche, as they always have been, there's more to computing space than low-end desktops and fleet shitbox servers

>>56365001
Confirmed retard.

Try to buy new hardware running HP-UX.

>>56365196
>Try to buy new hardware running HP-UX.
...like these? http://www8.hp.com/us/en/products/integrity-servers/

wow that was really hard

>>56365296
Said server but meant workstation.

Also
http://www.infoworld.com/article/2609688/unix/ibm-s-losing-ground-with-unix----and-oracle-may-follow.html

Unix Server sales down over 30 percent and falling.

you should literally only be running GNU these days

>>56365376
>Said server but meant workstation.
unix workstations are dead my man, and HP-UX like AIX was never really a big player in that mostly Sun and SGI-dominated market in the first place

>Unix Server sales down over 30 percent and falling.
it's been that way for years, doesn't really mean it's going away entirely, the same way that mainframes aren't dead just because microcomputers killed time sharing

the SysVs and their exotic hardware for now still have their place in niche applications, mostly "mission critical" services and big databases

>>56365402
go home dick you're drunk

>>56365459
>you're drunk
You're just jealous because I'm fucking right

>>56365459
>it's been that way for years, doesn't really mean it's going away entirely,
That's pretty much THE definition of legacy. You confirmed it by bringing up mainframes in your argument.

>>56364134
/thread

>>56365469
>Linux is the only option
kek, sure thing kid

>>56365469
maybe
desu though GNU stuff gets the job done but everything has its strengths and weaknesses aside from the basics of software compatibility

diehard unix shills also like to shit all over gcc because it encourages shit programming practices by being a little lax with warnings and compile time errors

>>56365503
>That's pretty much THE definition of legacy.
since when the fuck have sales and market share been "THE definition" of legacy? are PCs legacy equipment now too because of their decline over the last few years? or macs because apple doesn't give a fuck about them while the iPhone continues to print money? that's absolutely ludicrous

these platforms are still actively developed (and developed for) both on the software and hardware sides, shit, oracle just fucking announced a new line of SPARC systems a few months back and HP is probably going to port HP-UX over to x64 for their new integrities before the decade is out, that's hardly indicative of a "legacy" platform

>You confirmed it by bringing up mainframes in your argument.
but those are actively developed and maintained too???

>>56365658
>mainframes aren't legacy.
k

>>Unix is actively developed.
Meanwhile HP-UX has been at 11x for decades.

>>56365658
>HP/UX on AMD64
I'm ready.

>>56365708
Meanwhile OS X has been at X for 15 years!

>>56365766
But you actually see advancement in OS X.

>>56365766
OS X's 13th version is coming out soon.

>>56365708
>k
man you sure know your stuff don't you

>Meanwhile HP-UX has been at 11x for decades.
just like Mac OS has been on 10.x since 1999?
and Windows NT on 6.x since 2006???

what kind of transcendent straw grasping is this??????

>>56365801
Read garner reports and actually spend time in enterprise datacenters. Unix is dead. Your "niche market" is euphemism for "legacy." Look at college institutions and you will see dwindling server racks with Unix servers and a huge increase in commodity x86 running RedHat.

>>56365529
BSD can be runned with GNU and so can MINIX 3

>>56365937
Most commercial Unix has more GNU/free software utils than commercial. Run Solaris today and it'll come with GNOME, xorg drivers/servers, OpenSSH and gcc...etc

>>56365937
No shit, what's your point?

>>56365865
>b-but they aren't filling entire datacenters anymore!
literally who the fuck cares
good on them, you don't need an exotic high-end system to run a mail server anymore, that means fuck all

you can still buy HP-UX, AIX and Solaris hardware and software straight from the original vendors with full support and regular software updates

there are still sizeable communities of users maintaining and developing new software on these platforms

it isn't a deprecated product, these platforms aren't kept alive solely for running ancient code, it isn't legacy no matter how many memes and fallacies you continue to regurgitate

>>56366133
>it isn't legacy no matter how many memes and fallacies you continue to regurgitate
Your memes/fallacies are gartner reports.
https://www.gartner.com/doc/3220521/vendor-blink-battle-unix-viable

Look, I'm trying to help you. it's time to retool yourself or you'll become as legacy as your OS choice.

Sun made all their money selling Sparc hardware, so they put out the word that Solaris x86 sucked. (Even though it was basically fine.) So when everyone started switching to x86, they didn't even look at Solaris and went right to RedHat etc.

GNU/Linux distros had a better user environment too, Solaris was stuck in 1992 for a long time.

File: 294291853.png (5KB, 248x61px) Image search: [Google]

5KB, 248x61px

>>56366212
>just read this summary of a source you can't read in full! it says legacy in it! see??
yeah ok
once again, fuck all of a point

>Look, I'm trying to help you. it's time to retool yourself or you'll become as legacy as your OS choice.
you know that one can be interested in and know how to use multiple platforms at once, right?

>>56366350
>ALL this unsurmountable evidence is fuckall!
You're a dense fucker aren't you?

Unless you're running a legacy ERP with an Oracle DB backend you have no need for commercial Unix. This isn't my opinion it's the opinion of the industry.

>>56363322
No GPL license.

>>56366431
>>ALL this unsurmountable evidence is fuckall!
what insurmountable evidence??? one half of a paragraph and clickbait from the same group of retards who predicted x86 would be replaced by the same stuff you're shitting on now 20 years ago?

>it's not ME saying this, just listen to the Industry™!
the same vague "industry" that's still making and selling these systems?
the same vague "industry" that's still buying them and using them and developing on them?

you're giving nothing but a bunch of bullshit "trust me these guys know technology and they aren't doing that" anecdotes and repeating the same point over and over again that you can counter just by going to the fucking vendors' web page and looking at their products

Oracle bought Sun is what went wrong.
Sun should have never been acquired.

>>56366643
>retards who predicted x86 would be replaced by the same stuff you're shitting on now 20 years ago?
That was the era of NT and the p6 pro processor. You're showing your ignorance once again if you didn't see all the reports that the Pentium Pro was the risc killer and that all science/engineering software should be ported to NT. Which all happened as predicted.

>>56366325
Solaris x86 was kind of lame though, pretty much the only thing it was good for other than Slowlaris-isms was maybe development

>>56363322
Nothing

File: make Solaris GPL.png (580KB, 1279x837px) Image search: [Google]

580KB, 1279x837px

Use reason, petition to make Solaris GPL.

Or else, let it die.

_ = require('lodash')

>>56366850
GPL is shit.

>>56366737
>That was the era of NT and the p6 pro processor.
hell no, NT was still 3 years out when pure RISC started turning heads, around 1989-1991 it feels like every other magazine columnist was jerking it to the "revolution", check out some old magazines from around that period on Google or the Internet Archive some time, it's a laugh

> if you didn't see all the reports that the Pentium Pro was the risc killer and that all science/engineering software should be ported to NT
there was a lot of hype around the Pro but in the grand scheme of things it was still an entry-level chip shipping with systems hobbled by the clusterfuck of the PC architecture, not to say it didn't make very important headway into that market and set the stage for its dominance later down the road

>>56366850
don't even think it's legally possible, the CDDL wasn't "just because"

>>56366590

RMS, plz go.

>>56366905
You said 20 years ago, faggot. That would be 1996 - 1997 the NT4.0/P6 era. Are you that fucking old that you think 20 years ago was the late 80's?

File: 1472534502597.jpg (60KB, 640x640px) Image search: [Google]

60KB, 640x640px

>>56366967
>taking a casual estimate that barely related to the discussion in the first place literally and jumping all over it
are you a shitposting algorithm or something?

>>56367012
>barely related.

NT4.0 and intel was the beginning of the end of Unix/risc. Add GNU/Linux and you see why Unix is legacy.

>>56363322
old people

>>56367060
x86 is considered cancer by pretty much everyone.

>>56366743
>Solaris x86 was kind of lame though,

Solaris OS was way ahead of Linux in the relevant period. The userspace sucked though.

If Sun had brought over apt or rpm and pushed Solaris x86, most enterprise stuff would have stayed on Solaris. RedHat got those early customers because Sun gave them away.

>>56367060
>>barely related
it was barely related though, it was just an off-hand remark about how tech news isn't some kind of indisputable gospel, specific dates don't really matter

>NT4.0 and intel was the beginning of the end of Unix/risc.
P6? sure, it was still low-end but it set a precedent that Intel wasn't going to settle for the pleb end of the market anymore, and it paid off soon enough
NT4? eh, not really, it was modern and really nice compared to 9x for low-end workstations, but it didn't have the software, stability or power that the SysVs had at the time, GNU/Linux was really what sent them packing

>and you see why Unix is legacy.
I still really don't see at all why you keep saying this for reasons other than "but it's old"

>>56367373

>NT4? eh, not really, it was modern and really nice compared to 9x for low-end workstations, but it didn't have the software, stability or power that the SysVs had at the time, GNU/Linux was really what sent them packing

Tell that to all the cad/cam software makers that quickly ported their wares to NT4 instead of supporting 8 different flavors of Unix. I remember a time when you could choose NT4 or Redhat on SGI PII's and DEC alpha machines.

>>56367373
>P6? sure, it was still low-end but it set a precedent

Pentium Pro 200 beat the outgoing DEC Alpha in SPECMarks.

Sure, there is nine ways you could call bullshit on that benchmark, but the marketing point was made. There was no reason to spend $$$$$ for RISC when Intel was nipping at their heels. Intel would win, and eventually they did.

And when NT4 came out, most or all of the UNIX workstation software market just quit. Stop talking to your 1996 self and look at actually what happened.

>>56367499
>Tell that to all the cad/cam software makers that quickly ported their wares to NT4 instead of supporting 8 different flavors of Unix.
CAD was pretty much the only market it did have any real success in, and even then, nobody was dumping Unix support for it, the likes of Catia and Pro/E stuck to Unix until the very end

IRIX's extensive library of exclusives definitely didn't make it over in full, probably because NT4 systems were still absolute trash for that kind of work

> I remember a time when you could choose NT4 or Redhat on SGI PII's
you mean the Visual Workstations that were hailed as the beginning of the end for SGI because nobody wanted to pay $20,000 for what was effectively an O2 that couldn't run any software you bought an O2 to run?
and let's not even get me started on how absolutely fucking shit Red Hat is on a VW

>and DEC alpha machines
that were also failures because they had practically zero native software options and spent 90% of the time emulating 386 code slower than an x86 system at half the cost?

just because it exists doesn't mean it was actually good

>>56367612
>Pentium Pro 200 beat the outgoing DEC Alpha in SPECMarks.
yeah... no
here's an Alpha 21164/300 https://www.spec.org/osg/cpu95/results/res96q3/cpu95-960722-01084.html
here's an Pentium Pro/200 https://www.spec.org/osg/cpu95/results/res96q2/cpu95-960610-00948.html
pretty sure the alpha isn't even the fastest that was available alongside the Pro/200 either

hell, even if the Pro did match up, Spec benchmarks are a shit metric of comparison anyway, since they ignore other critical details like memory bandwidth, even if a given Pro could theoretically hit levels of performance that could match a low-end 21164 or an R10K, it doesn't matter for shit when the shitty slow bus design is starving it

> There was no reason to spend $$$$$ for RISC when Intel was nipping at their heels.
now this I can totally agree with, the Pro, as ultimately unimpressive as it was and with all of its flaws, still paved the way for a new generation of commodity workstations more affordable than ever, and the big vendors were foolish not to nip that shit in the bud when they had the chance

>And when NT4 came out, most or all of the UNIX workstation software market just quit. Stop talking to your 1996 self and look at actually what happened.
like what? D'assault didn't quit, Wolfram didn't quit, Softimage didn't quit, Alias didn't quit, Discreet didn't quit, PTC didn't quit, you could go on forever. I could see maybe small-time developers making the switch, but for the big players it was just another platform to support, not the only one.

>>56367612

Not to mention that AMD snatched away ALL THE ENGINEERS from DEC (well not all, but a significant portion) that designed the Alpha.

And then Microsoft convinced HP to dump Alpha for Itanic, which doomed Alpha forever.

Because AMD was first to market with an affordable 64 bit cpu that didn't make you throw out all your binaries if you were coming from x86.

I liked Alpha. It's because of bone-headed management decisions that it's dead.

I'm still not over Carly killing the calculator division. /goes to cry in corner

>>56367881
>And then Microsoft convinced HP to dump Alpha for Itanic, which doomed Alpha forever.
HP were the ones who spearheaded and developed the damn thing, they weren't convinced, they were the ones doing the convincing.

>>56367794
get educated from the past.
https://wohlersassociates.com/feb96.html

>>56367870
>D'assault
It's just Dassault.

>>56368090
>Although some of the newer packages, such as TriSpectives from 3D/Eye and SolidWorks 95 from SolidWorks, were developed from the ground up to run on Windows, most of the Unix packages that have been ported over to Windows were not changed considerably to conform to Windows standards.
>If high-end CAD software developers want to succeed in the Windows market, they will need to create Windows products that look and behave like other Windows products. Until they do, customers will not embrace the Windows version of their products and the transition to the Windows environment will appear slow. Some may blame this slow migration on the Windows platform itself, although the real problem may be with the poor effort put forth by the software vendors.
>But does that mean Unix as a platform for CAD applications will die eventually? Quite possibly. There will come a point in time that software developers can no longer justify the cost of developing and supporting individual Unix ports. With such a small volume, it won't make economic sense. But this won't happen soon, and this is why: customers of Unix hardware and software have invested large amounts of money in their systems and in training their people. Those who were a part of the decision to buy Unix will be slow to admit that Windows might offer a price/performance advantage over an operating environment they helped create.
what are you trying to prove to the contrary of what I posted here? it's literally what I said, the article mentions a few small vendors embracing NT and calls out big vendors for not giving a shit and providing either half-assed ports nobody wanted to use or absolutely nothing at all, nothing in there says anything about this mass exodus of Unix developers to NT4 and P6 systems

>>56368147
don't know why the fuck I spelled it like that, my bad

Solaris is still alive at version 11 point something. The kernel and underlying technologies are as UNIX as can be. But since Sun got out of the workstation market long before they themselves went down, using it as a desktop OS is a little meh. So it's basically a UNIX with your standard GNOME desktop and GNU utilities added. Solaris 11's strengths are in the server and enterprise.

>>56368223
Shit, even while they were still making workstations, it felt a little lacking towards the Ultra revival period.