I was interested
I rather fancied one of these wee beasts. Not that I have a great reason to have one beyond fleeting nostalgia. I'm really glad I passed now, but I feel bad for the folks who tipped money into this burning pit.
962 posts • joined 18 Oct 2010
+1 purely because I have also fought the good fight in the PC World tech department. Wasn't it fun when a family brought the computer in for repair and you could see how sheepish the eldest son looked because he knew there were lots of pictures of naked girlies in a hidden folder? And he knew that you knew.
Packard Bell - that's a name I've not heard in a long time... A long time.
One of the comments above send my mind down a rabbithole. How feasible would it be, with metal 3D printing, to set the device up in a vacuum, and print (say) a sphere with internal bracing against collapse? Then, could you flood the chamber with air and keep a balloon full of vacuum which would float?
And then what use would it be? I guess you could add buoyancy to all sorts of things, but would *lots* of buoyancy be better than just using a lightweight gas?
Something I've pointed out to many people over the years. There may be more than one way to skin a cat, but given a specific level of technology and a given physiology of cat there's only one optimum way to skin a cat.
Ergo, engineers trying to strive for the same (difficult) goal will often tend to converge on similar designs even when working in isolation.
There are two ways to do it. You can either have Kodi launch an external application for Netflix (which is hardly in the spirit of the thing), or in the v18 builds there's a new feature called "inputstream.adaptive" which allows plugins to run Amazon Prime and Netflix.
v18 is still in development, though, so you're on the potentially unstable nightly builds for that. Give it a little time until Leia is released, and you should be good.
There's a BBC iPlayer plugin for Kodi. I think it's called WWW iPlayer, so it's in the wrong place in the download list. I run it on my Pis. Can't check just now because the kids are watching The Crystal Maze, and I can't be arsed going upstairs to look at another :P
I believe there's a mechanism for Netflix and Amazon that doesn't require Windows, but I've not looked into it in any depth.
How about ZFS or another filesystem with snapshotting built-in ?
The problem is with applications keeping the files open / buffers needing flushed.
VMware has a button on the snapshot window to "quiesce guest filesystem", which forces it to flush the buffers, but won't write any application state out to the drives. In general you'll get away with re-running transaction logs to get databases (including Exchange Server) up-to-date. Personally I never ran into a problem with it (a bit further down the road I deployed file-level backups using Bacula which meant that those were closed, flushed and safe), but there are certainly situations where it would be a Bad Thing™.
Did I use it in the past to cover backups for several small businesses? Yes, I did. Was there ever a problem restoring files? No, there was not. Would I use it for a server handling hundreds of users? No, I wouldn't.
Snapshots at a hypervisor level have never been a backup mechanism.
Now snapshot-aware backups are of course useful, but that's not the same thing.
How about home-rolled Perl scripts that kick off a snapshot to freeze all of the data in the middle of the night, and then extract the frozen copies of the drives to a staging JBOD before removing the snapshot and committing to tape?
Worked a treat for years in a variety of guises, covering VMware GSX, VMware Server 1, XenServer 5.5 and finally ESXi 4.1. Especially when no money was forthcoming for server software, so VCB wasn't an option. Much happier to be using Veeam these days (although I've just had an email saying they're putting their prices up...)
...Just the odd VM here and there. I did run a Citrix XenServer some time ago. Turns out that taking snapshots for backups was a bad idea. ISTR it was all LVM snapshots, but XenServer never actually removed them - just took them out of the admin console. The whole thing got slower and slower as the backup chains got longer, and it took up more and more SAN space. Eventually (I suspect after 256 snapshots), it just refused to snapshot any more.
That was the point where I finally got funds to put in VMware ESXi 4.1, so that shows you how long ago it was. I sincerely hope it's improved since then.
The Xen Hypervisor itself was pretty solid though!
"If you thought that, then you've not been paying attention."
You got that right! I've only seen the odd video from Blue Origin - things like their capsule landing test. Whilst these things have been impressive, SpaceX have managed to out-spectacular them each time.
You're dead right that there'll be plenty of room in the market for both players. And that the market will grow significantly as launch costs drop. The people who should be most worried about this will be the incumbents, like Arianespace and ULA. Maybe this will help get some of the political pork out of spaceflight, and allow NASA to spend their funds with the lowest bidder.
"ie over 3.3 times the thrust for less than 1.6 times the payload. Why so little increase in payload? are the figures wrong?"
The figures don't greatly surprise me, to be honest. The payload is the smallest part of the package.
If you want to carry more payload you need more engine. And then you need more fuel for both the bigger payload and the bigger engine.
Then you need more fuel to carry the fuel. And more fuel to carry *that* fuel. And more fuel to...
You get the idea - it's turtles all the way down. Eventually somebody round up a decimal point and it becomes enough. But even small increases in payload mass can have a remarkably distorted effect on launch mass.
I've been considering Blue Origin as the poor cousin between them and SpaceX. After all, Blue Origin have been doing hops, whilst SpaceX have been orbital for years. This ups the ante rather a lot though.
Still, Musk himself can sit back in his volcano-lair happy that he's changed the world however the race between him and Bezos pans out. It's not like he'll struggle to pay for cornflakes at any time soon!
Two years is a long time, though, and the Falcon Heavy is slated to fly three times in the next six months. Fun times for big toys!
A friend and I were reminiscing just last weekend about the 'click' when our Orchid 3DFX cards kicked in. That was when you knew something glorious was going to happen!
Besides that, there's now a Glide wrapper so you can force the 3DFX calls to Direct3D for newer cards. I was playing Dungeon Keeper 1 last night. :D
Thumbs-up for the Squeezebox.
I have a Squeezebox 2 and 3 Booms. Love them, and with the open-sourced server anyone can jump in and make a device. It'd be nice to be able to buy an off-the-shelf device that would work with it...
I know you can slap together a Pi with SqueezeLite, and there are half-decent DACs and amps you can hook into it, but a nice, neat turnkey box (like a Sonos) would be great.
You noticed that too?
I can confidently put my hand up and say that we have a tape library that didn't read a batch of labels. Some it would read inconsistently, and others not at all. Got hold of a label vendor and gave them the make/model of the library. "Yes, we'll provide labels that work." And they did.
Another On-Call that I've lived and breathed...
There's only so much spectrum to pass around.
Surely the sane urban solution is more towers at a lower power, handling fewer users? And for people to accept that eventually these things get full. Also for people with static applications to use a wire...
Aren't the smart electricity/gas meters supposed to report via GPRS? Because once they're rolled out they'll be in the field for a long time, tying up that spectrum and forcing the mobile phone operators to support GPRS for decades. That can't help - it'd be like having an 802.11b device permanently on your wireless...
"Although I agree Linux and Python are a bit of a mouthful to swallow at once compared to switching an 8bit on and typing "10 print 'hello world'"
That's why I'll be setting one up for my lad with RiscOS. BBC BASIC FTW!! And it's got enough poke to do big-boy stuff. And if he ever decides that's what takes his interest, there's an ARM assembler in the box too.
By the time he gets to that stage, a bit of Python and C shouldn't be beyond his grasp.
I'll even encourage my daughter to give it a bash. She has a very different mindset, though, so I'm not convinced it'll be her thing.
"Politicians didn't listen to the sneering experts"
There's a reason for that, you know. "People in this country have had enough of experts" - Michael Gove said so, so it must be true.
I'm just wondering where all the tax revenue is going to come from when the entire financial sector collapses due to an international loss of faith in the encryption of British comms...
"And that is why you should have an external system monitoring and alerting, not rely on kit on just one site."
And that is assuming there is the budget and the equipment to do this. Most company IT systems don't have that luxury. And a big part of the problem, for a long time, was BT - let's be honest. Given the cost of leased lines back in the day, it was cripplingly expensive to monitor client equipment.
In 2000 we had a guy come in to demo Unicenter TNG to us. Nice South African chap. He tried to push it that we could use this to monitor a remote system over a simple leased line. Then we explained to him how much leased lines cost in the UK...
Somewhat different these days, but it still assumes that you have multiple sites to play with.
The fibre run is Virgin, not Openreach. And you're right in what you're saying - when Openreach work they're (generally) great. But when they don't it turns into a game of pass-the-parcel with the problem.
I've had a DSL line problem bounced between ourselves and three other agencies for about 2 months before I walked onto the site with snips and a punch tool and reterminated it. Worked great after that. And I have had a number of episodes like that...
In the case of this Manchester site, though, I reckon there's water in the cables, and that would be a mammoth nettle to grasp. The client are delighted with their fibre line. And Pulsant were an easy choice because that same client have had SDSL and EFM from them for years beforehand with exactly zero issues.
(I don't work for Pulsant, and I accept that they've had their issues over the years, but their business comms have been pretty robust.)
One of our clients had super-shonky ADSL. I think the duct is flooded or something, because it's bad all the time, but really takes a crap in foul weather. FTTC, we though - that'll get us past some (lots) of the ducting intact. Exchange is live. Cabinet scheduled for <5 months. Then they pulled the plug on the upgrade so they can push out FTTP.
What? So we can't get FTTC in there this year and FTTP later? We ditched Openreach and got an all-fibre line in via Pulsant. Sorted.
Biting the hand that feeds IT © 1998–2019