Re: Scamdemic
With colossal global interventions Covid was about 10x worse than a bad flu year. Without those interventions it could easily have been an order of magnitude worse again.
48 publicly visible posts • joined 28 Jan 2011
I've seen Word documents turned into a heap of slop in Word. If you're handed files from a different person, or running a different version of Office, or a different version of Windows, then chaos can ensue. LibreOffice isn't necessarily going to give a perfect rendition, but I usually see it do a pretty good job.
The idea that CDDL was chosen to be deliberately incompatible with GPL is simply untrue.
Sun had to persuade a long list of existing licence holders and intellectual property owners to agree in order to open-source Solaris. CDDL was the compromise that resulted. (It also had to be a per-file license, rather than a per-project license, as lots of files in Solaris/OpenSolaris/illumos have different licenses.)
We wouldn't have expected Linux to simply take the code for, say, ZFS, and simply recompile it. A cleanroom implementation of the format seemed more likely, and it's been a bit of a surprise that ZFSonLinux decided to import the existing source and create a porting layer.
We had an east coast datacenter in the path of Hurricane Sandy.
Manglement were quite keen to keep it running, as we had an important contract. (The "see inside" feature for books on a certain major online retailer was served from there, although it got relocated very shortly afterwards.)
We had enough diesel for about 24 hours. We sent extra staff over, so we had someone on his Harley riding around local gas stations, depots, whatever might have fuel trying to get more diesel to keep us going a bit longer. Not a drop. After a while the local law enforcement caught up with him and rather strongly suggested that the curfew that was in force ought to be obeyed. That service was offline for days.
Based on the ability of the office suites used by people who my other half exchanges documents with to read the ODF format (essentially, zero) I wouldn't be so sure about claiming success. Yes I've heard that in theory Office can read the things, but in practice all that happens is complaints about unreadable documents.
One thing I was conscious of at FOSDEM was that while there are a lot of us of the (slightly) elderly persuasion, overall most attendees were quite a lot younger than myself.
I'm starting to suspect that the problem isn't really bringing in younger developers as such, it's bringing any developers at all into a mature project. So, over time, mature projects tend to stick with the same core of contributors, which naturally ages.
For something that's not an especially large estate, requiring a "dedicated team comprising more than 150 OpenStack experts and 50-plus Kubernetes administrators" indicates a significant level of complexity. That tells me that deployment needs a team of dedicated trained experts, which tells me that the technology isn't suitable for deployment by the majority of organisations.
And also those staffing numbers tell me it's 3 times worse than Kubernetes. Gulp.
(If a disinterested partner contributes over 5 million lines of code then maybe, just maybe, the suite is far too big and bloated.)
About the only issue I had with PostgreSQL is the upgrade mechanism - across a major release, it's dump and restore, which involves downtime, and is a right pain.
Sure, there are ways to avoid it, but they're relatively advanced, and need to be set up before you start; they're not easy to retrofit.
If the project provided a built-in transparent upgrade path, I'm sure users would upgrade more readily.
The GPL wouldn't make any difference - indeed, using the GPL is quite explicit that you cannot stop anyone charging for your software. (The only thing about the GPL is that you require modifications to be published, but even that doesn't apply to services hence AGPL.)
Many were. Many fell down.
Until fairly recently in human history much of engineering and architecture was of the try-it-and-see variety. There's massive survivorship bias because we only now get to see the ones that worked. Hopefully these fields have learnt the hard way what is necessary for success.
Solaris (and thus illumos) have had boot environments in their current form (ie, on ZFS with snapshots) for about 15 years or more. You can have as many as you like (within reason, and within the arbitrary limits of grub when that was the bootloader).
Before that, certainly in Solaris 8 through early version of Solaris 10, we had the equivalent of the 2-partition setup with Live Upgrade. Which did work, but was clumsy and inefficient compared to the fully integrated and instant snapshot environments we're used to now.
One thing that strikes me is that development effort seems to be aimed towards random fiddling, wasting time adding features users don't want or removing features that users like. Looks like some leadership required to bring back some concentration on their core competencies.
So, they're taking the same path that Linux distributions like RHEL/Fedora and Suse have already taken - bumped the Microarchitecture level to x86-64-v2 or so.
(Yes, I know Windows doesn't think about hardware support in terms of Microarchitecture levels the same way, but that's what it effectively is.)
At least in the open source world you can jump to a different distribution, or build it yourself. Although the pressure to bump the required baseline is building across the board, as more and more applications start to require features like AVX.
It's "illumos". No capitals. Some of the team get quite picky.
But yes, illumos continues the "just works without randomly breaking stuff on a regular basis" philosophy. Like Solaris, sits quietly in the background doing its job without requiring an army of squirrels constantly caring for it. Binary compatibility with Solaris 10 is excellent.
Used for new and very cutting edge stuff, like Oxide and Helios.
(Some of us are still running illumos on sparc. Now that is a bit niche, I'll admit.)
Perhaps groping randomly through all the rubbish on the current web isn't the best - or even a good - way to find information.
Early on, there were a number of excellent sites that provided curated lists of quality content. While that model doesn't solve all the use cases for search, it did do a much better job of some of them.
"OpenSolaris did the same back in 2007, and it's doing fin— Oh wait."
The in-kernel smb server is still present in illumos, and actively maintained.
The interesting part of that is not so much the SMB server part, but that the use of Windows SIDs is fully integrated into the OS and ZFS, so you can manage ownership and permissions for Windows users natively.
"Applications just use so many dependencies, it boggles the mind."
And yet rather than actually solve the problem, the industry seems happy to encourage ever deeper dependency trees and throw money at attempting to handle the inevitable fallout.
What happened to doing proper software engineering?
While Pale Moon was well behind the curve for a long time, it's recently seen some major improvements and now works well enough for the most part. Whether it can keep that up or fade away again remains to be seen.
But it's a constant game of whack-a-mole trying to handle all the new misfeatures that all the chromesites have implemented.
While SPARC will be removed from upstream, that isn't quite the end of illumos on SPARC. Tribblix will simply fork illumos and continue with it's illumos on SPARC distribution from what will then essentially be a static base. Makes life easier on both sides - illumos mainstream can evolve more aggressively, and those of us using unchanging SPARC hardware can evolve much more slowly.
I don't think the divergence has gone away, but I do see the emergence of Cults (the Cult of Cloud, the Cult of Containers, lots of Cargo Cults) where people live in an echo chamber and largely interact with other cult members who share the same views, and who believe that there isn't a world outside the cult. Which mirrors trends in society overall.
Opening up the Solaris source (again) would make life very ... interesting ... for illumos.
We've diverged quite a bit, and have enough changes of our own, to the point that pulling in changes would actually be quite difficult. Hardware support for current SPARC models would be good, but I'm not sure there's all that much else.
Not terribly significant, of itself.
There are people using MongoDB on Solaris/illumos; they're likely using their distro packages or building it themselves. And by the sounds of it, not paying a support contract. So those users become invisible.
A bigger problem is the relentless "everyone's abandoning Solaris" message being spread, which will tend to be self-fulfilling.
You can't read anything into the activity of the solaris-userland repo - that's just the holder for the external open-source projects they ship as part of the product. Activity there is largely driven by CVEs and releases of those external projects, and it may not be what they're using as the primary internal repo in any case.
But yes, Oracle's stewardship of a flagship product has hardly been encouraging. You want that stuff, there's always the illumos project.
It seems a little odd to call the CDE calendar server problem a Big Red boo-boo. That code comes from *way* back.
Besides, nobody competent would actually have that running. Although I suppose that leaves a fairly big opportunity.
The Java scores look scary, but the list and scores aren't terribly helpful. Apart from telling you not to run Java in the browser, which is a message people should have gotten many many times already.
Oracle's storage offerings are even more limited than the article suggests - as far as I'm aware the 2500 series is no longer offered.
Equally damaging is the removal of the X4500/X4540 line, which didn't really have much competition. There's no direct attach storage either. This is particularly odd given how well such storage plays with ZFS.