Linux Mint 14 was released yesterday. After looking at it, I have no plans to upgrade.
To start with, there are no updates I find compelling. Yes, MATE has been updated, with several bugfixes they say are long overdue. I guess they don't bother me enough to rate poking at. Beyond that, though? Nothing jumps out at me.
I'm also told by more than one person (starting with my roommate) that the version immediately after an LTS of Ubuntu is usually the worst choice possible, since many upgrades are put on hold for the LTS version, and so a lot is added to that next release of dubious quality and stability. Mint 13 is an LTS release, and I'm just as happy about that.
I'm also a bit dismayed that the recommended upgrade process for Mint is a fresh install and reload of your personal data. I understand their viewpoint, but come on, guys. This is one more thing Apple gets right. I've never had problems with updating OS X from major release to major release using the OS updater, and I've been doing it since upgrading from 10.1 to 10.2 - that is, for over a decade.
A fresh install runs a major risk for me of missing something I added. Yes, they do allow backing up and restoring the package database (and, I presume, automatically installing everything that was restored), but not everything is in the package database. There are a few packages I've installed that are newer than what's in the repository, usually because they've had major improvements since that version.
Then there's the Firestorm build stuff. Most of that is either in packages, my home directory, or /usr/local/, but I'm not 100% sure of that. I'd just as soon not reinvent that wheel if I don't have to.
Since Mint 13 is an LTS (long term support) release, I'm not worried that there will be a security fix that I won't be able to get without upgrading. If they prove me wrong on that point, I'm not going to be pleased.
So I'll be on Mint 13 for a while yet.
Wednesday, November 21, 2012
Wednesday, November 7, 2012
Fingerpointing
While I wait for the replacement memory riser card to arrive in the mail, I've been using the Linux box for more than just running Firestorm. I'm getting used to things, at least enough for that purpose and some limited web browsing and such. I'm running into the next hurdle, though: What do I use for a pointing device?
To understand my issue, you need to see just what the workspace I sit in front of for hours a day looks like:
Quit snickering. Yes, I get real work done here.
The screen on the left is the Linux system, and the one on the right is the Mac Pro. The black keyboard is one I got with a refurbished HP desktop I bought for a firewall; it's the cheapest piece of crap generic keyboard I've ever seen. I typo on it at far higher a rate than I do on anything else I've ever used. The keyboard on the right is an Apple wired keyboard I love. To the right is an Apple Magic Trackpad, sitting on top of a Wacom Graphire tablet. The trackball in roughly the center of the picture is for the Linux box.
I love the Apple keyboard and Magic Trackpad. The Magic Trackpad, in particular, enables multitouch gestures - and I've become very dependent on that. It's far too easy to get used to flicking two fingers to the left to back up a page in the browser history, for just one example.
The one capability I miss most, though, is the ability to scroll by dragging two fingers in any direction. I've been used to scrollwheel mice for years, and the trackpad's scroll functionality is an ingrained habit. The trackball doesn't have it, and it's a real annoyance.
As you can see, a mouse is Right Out. There's no place to use one, let alone two. I also gave up on mousepads a long time ago, and there's a space worn in the finish of the desk under the Graphire where I used mice without one for a few years. I'm not fond of the trackball, either; it makes some operations either hard or impossible.
So I want a multitouch trackpad for the Linux box. Unfortunately, there's no one place to go for information on support for multitouch input for Linux. There's one page that has some information, but it's incomplete, and then it refers the user to another page that is a firehose of information without actually being informative. I just want one place that says "this trackpad works this way, and this trackpad works this way", so I can make an informed buying decision without having to become an expert on which device does what.
I think the Magic Trackpad actually works for multitouch on Linux, but I'm not sure. I may find myself trying the one I have when I can have the Mac down for a bit. If so, I'll get a second one (and be damned to the Apple haters out there). I'm already pretty sure that the piece of crap keyboard is getting replaced with an Apple keyboard just like the one I have.
Why do things have to be so hard, though?
To understand my issue, you need to see just what the workspace I sit in front of for hours a day looks like:
Quit snickering. Yes, I get real work done here.
The screen on the left is the Linux system, and the one on the right is the Mac Pro. The black keyboard is one I got with a refurbished HP desktop I bought for a firewall; it's the cheapest piece of crap generic keyboard I've ever seen. I typo on it at far higher a rate than I do on anything else I've ever used. The keyboard on the right is an Apple wired keyboard I love. To the right is an Apple Magic Trackpad, sitting on top of a Wacom Graphire tablet. The trackball in roughly the center of the picture is for the Linux box.
I love the Apple keyboard and Magic Trackpad. The Magic Trackpad, in particular, enables multitouch gestures - and I've become very dependent on that. It's far too easy to get used to flicking two fingers to the left to back up a page in the browser history, for just one example.
The one capability I miss most, though, is the ability to scroll by dragging two fingers in any direction. I've been used to scrollwheel mice for years, and the trackpad's scroll functionality is an ingrained habit. The trackball doesn't have it, and it's a real annoyance.
As you can see, a mouse is Right Out. There's no place to use one, let alone two. I also gave up on mousepads a long time ago, and there's a space worn in the finish of the desk under the Graphire where I used mice without one for a few years. I'm not fond of the trackball, either; it makes some operations either hard or impossible.
So I want a multitouch trackpad for the Linux box. Unfortunately, there's no one place to go for information on support for multitouch input for Linux. There's one page that has some information, but it's incomplete, and then it refers the user to another page that is a firehose of information without actually being informative. I just want one place that says "this trackpad works this way, and this trackpad works this way", so I can make an informed buying decision without having to become an expert on which device does what.
I think the Magic Trackpad actually works for multitouch on Linux, but I'm not sure. I may find myself trying the one I have when I can have the Mac down for a bit. If so, I'll get a second one (and be damned to the Apple haters out there). I'm already pretty sure that the piece of crap keyboard is getting replaced with an Apple keyboard just like the one I have.
Why do things have to be so hard, though?
Friday, November 2, 2012
Jumping faster than planned
My hand may be forced sooner than I'd wanted. The Mac Pro has started getting flaky on me. Twice in the past couple of days, it's locked up with a completely white screen. It's had a problem for a while with seeing all 12 GB of memory in it; most, but not all, of the time, it's only seen two 1 GB DIMMs where there are 2 GB DIMMs.
I'm suspecting that something in one of the memory riser boards is failing. There are two of them in the machine, with two 2 GB DIMMs and two 1 GB DIMMs each. I may try replacing one, just to see if that cures the problem, but first, I'll pull it apart and blow the dust out of it. Sometimes the little steps are the best.
If it stays flaky, though, I'm going to have to make a decision, fast: do I try to replace it now, or do I take the plunge? I'm not ready to go Linux whole hog yet. There are too many things I depend on OS X to do right now, from generating Microsoft Office documents to using Citrix Remote Access software to using Microsoft Remote Desktop over Cisco VPNs to doing GoToMeeting meetings, all for talking to customers and keeping the bills paid. If I drop OS X, I need to have an alternative, and that alternative had better not be "run Windows!". If I wanted to run Windows, I'd be doing it already. I can run Windows 7 in a VM, but if I spend all my time in the VM, then I might as well run Windows for real and do away with the middleman.
And no, not doing any of the things I listed in the last paragraph is not an option. I'm not in a position to tell customers that I'm not going to work with what they demand of me.
Replacing the entire box will cost me a kilobuck, but in the process I'll get an upgrade by one generation to one that can run Mountain Lion natively. That's a kilobuck I just don't have at the moment, however.
So, how do I keep serving my customers in a Linux environment?
I'm suspecting that something in one of the memory riser boards is failing. There are two of them in the machine, with two 2 GB DIMMs and two 1 GB DIMMs each. I may try replacing one, just to see if that cures the problem, but first, I'll pull it apart and blow the dust out of it. Sometimes the little steps are the best.
If it stays flaky, though, I'm going to have to make a decision, fast: do I try to replace it now, or do I take the plunge? I'm not ready to go Linux whole hog yet. There are too many things I depend on OS X to do right now, from generating Microsoft Office documents to using Citrix Remote Access software to using Microsoft Remote Desktop over Cisco VPNs to doing GoToMeeting meetings, all for talking to customers and keeping the bills paid. If I drop OS X, I need to have an alternative, and that alternative had better not be "run Windows!". If I wanted to run Windows, I'd be doing it already. I can run Windows 7 in a VM, but if I spend all my time in the VM, then I might as well run Windows for real and do away with the middleman.
And no, not doing any of the things I listed in the last paragraph is not an option. I'm not in a position to tell customers that I'm not going to work with what they demand of me.
Replacing the entire box will cost me a kilobuck, but in the process I'll get an upgrade by one generation to one that can run Mountain Lion natively. That's a kilobuck I just don't have at the moment, however.
So, how do I keep serving my customers in a Linux environment?
Subscribe to:
Posts (Atom)