Your Usenet post describing the first public version of the Linux kernel compared and contrasted it with GNU, which itself was designed to be very Unix-like. Did you have a "final vision" for Linux in your head in the early days? I mean that as, did you see Linux getting to be a suitable Unix replacement that was still very Unix-like and then stopping, or did you have some sort of vision of a radically different operating system built on top of your kernel?
Obviously, you use Linux every day. What distribution do you use? Do you have a preferred desktop environment or window manager? A preferred text editor?
Australia has corporate tax rates that are in general lower than those in, say, the United States. The US has lower tax rates for corporations with income less than $100,000, but I would very much assume that this studio made more than that. Regulations for this sort of industry are essentially the same around the world as well.
The cost of doing business in Australia is negatively impacted because of major time zone differences from other English-speaking nations, and the significantly higher costs of transportation to/from and telecommunications with what is really a very geographically isolated nation.
Hah. On the Windows 3.1 systems at my high school I would change the screensaver message to something like "FUCK THA POLICE" or whatever and then use the ATTRIB command to mark WIN.INI as read-only, meaning it was impossible to change the message back using the UI.
I have not seen a laptop in close to a decade that didn't work 100% out of the box with Linux. Ubuntu on my laptop actually supported more of the hardware than OEM Windows did without having to find third-party drivers on some website. I remember having to manually configure my sound card with Linux, but I don't remember the last time I did it; it's been at least a decade if not more.
So yeah, the old objections to Linux are, to use an appropriately old term, FUD.
The main thing I'd like to see for 4.0 is a massive simplification of the kernel, removing features that are no longer used anywhere. There's a lot of duplicated functionality in the kernel - two different ways to report hotplug events, three different ways to report ACPI events, several dozen filesystems, some of which don't actually work or even have userland tools that will compile on modern kernels.
So I'd like to see a winnowing of kernel features, down to a saner, all-known-to-work set.
Also, can we please at some point make/proc a bit more sane? Why do I write kernel configuration parameters to files in proc? Why not to sysfs or configfs?/proc, IMHO, should just hold process information. It's confusing.
I would argue that there should be some sort of regulation that ensures that phones are interoperable with one another for "phone stuff". That is, if you sell a phone in this country, by law it should be able to make a phone call to every other phone sold in the country. The problem is, what qualifies as "phone stuff" is rapidly expanding.
iMessage is a good example. Apple is trying to leverage its dominant market position to make text messaging something that's iPhone only. Remember the whole debacle with people who had an iPhone and then didn't suddenly not being able to receive text messages from other people who still had iPhones. Apple's solution was broken and only partially effective - and I think at least somewhat intentionally so. Same with FaceTime. You want to talk to your friends with an iPhone? Well, you need an iPhone too!
So yeah, we as a society need to decide what we define as "phone stuff". Having the ability to communicate with every other phone for "phone stuff" is critical from an economic perspective, and eventually will also be so from a safety perspective. Requiring inter-phone communications to be standardized isn't too far-fetched of an idea.
(Requiring the same non-phone-stuff apps to work on different platforms though is stupid.)
This sounds a little insensitive, but, don't be disposable. You're a Windows admin. Great. So are a million other people. If you're a Windows admin who also knows some programming, there are maybe 250,000 people with your skill set. If you add in that you know some Linux, maybe 100,000 people.
What I'm saying is, if you want to be safer than the average employee, don't be average. Enhance your skill set.
This isn't even my first Slashdot account, my old one dates from 1998 or 1999.
When I first got on Slashdot, there was meaningful, technical discussion. A good number of actual experts in scientific and technical fields were present. Yeah there were always trolls and people racing to have the initial comment, but I feel like the entire tone of Slashdot has changed. You rarely get technical experts on here anymore, the trolls are just as prevalant if not more, and the entire readership has turned a lot to the reactionary right - scientific stores get inundated with "but that's socialism!" comments. There's even a fair amount of junk science in the comments now.
I read Slashdot now solely for the headlines, primarily because I never got the hang of Reddit. The comment section has been basically useless for a while now.
There was the botched iOS 8 update that broke phones' data connectivity, and required lots of phones to require reinstallation via iTunes. Then the fix they released was broken and they needed to release a fix for the fix. There might have been a third fix, I can't remember.
Seriously, if my phone isn't capable of doing whatever task I need to do, it means I probably need my full-on laptop anyway. Add in the fact that a tablet either requires wifi or requires cell service but can't make calls and it becomes obvious why the market is behaving this way.
So over my nearly 20 years in IT/CS, I've seen a few:
I worked for a large retailer. We migrated from an old frame-relay leased-line network to a much more capable multihomed IP-over-VPN configuration to connect all of our retail locations around the country back to HQ. This new system worked well. Our CIO retired, and a new one was brought in. CIO Magazine a year or so later had an article about "Satellite Internet, The Future?" Our CIO then "spontaneously" started lobbying to get us to scrap our efficient, inexpensive, high-bandwidth network for a satellite system.
I can't tell you how many projects I saw rewritten in Ruby on Rails just because that was the new hotness, only to be abandoned later when everyone realized that Ruby is awful.
I myself wrote a bunch of stuff in Erlang not because it was the best language but because that was the new hotness.
Two unchanging things I've noticed are:
A lot of time, the new hotness makes common problems go away or common tasks easier, but ends up making more complex things harder. This isn't necessarily a bad thing, but people tend to get stuck in the model of thinking that the new technology has to be used for everything, and they end up shoehorning their complex projects into frameworks that aren't the best choice.
No matter what the new technology is, and no matter how fantastic it is, it's not going to replace C/C++ for systems-level work, and Python and Perl aren't going anywhere. Truly successful technologies have long tails.
I would genuinely appreciate any stories you have to share - not because I don't believe you, but because Slashdot has recently become so "angry white male" that it's depressing.
(I've been on Slashdot since 1998. I remember when article comments were a welcoming forum. Now it's yet another forum where "liberal" on its own is conisdered an insult and labeling someone as such immediately means they are no longer worthy of consideration.)
Greg Egan's stuff wouldn't translate well to the screen, I think. I absolutely love his work (Permutation City is one of my favorite books, and I loved Schild's Ladder, Quarantine, and all of his short fiction). The problem is that there's too heavy of a cognitive science/philosophical bent to them. You'd have to have a character sit down a monologue for a while to get everything out.
Okay, here's the deal. I am passionate about computer science and programming. It's what I do, both for my job, and as my only hobby. I write code for open source projects, and I write code for work, and I design little one-off projects for my own entertainment.
I stayed up all night every summer growing up teaching myself how to code. When I go to the used book store, I go to the section and buy old computer science textbooks talking about esoterica (I'm the only person I know under 45 who knows any APL, for example). My bedtime reading last week was the Oberon System manual that I got off eBay for $5.00.
All this was despite the fact that I grew up in rural Texas and got my ass beaten on a daily basis for being a "geek". The fact that my family was the only non-Christian family in town meant that I couldn't go to the school administration for help; when I tried it turned into a "let's pray for you, son." And yet, I kept doing it because I was passionate about it.
And guess what? If you're that passionate about something, you'll do it regardless of what your peers think. You'll *make* it happen. We didn't have any money growing up, so I'd stay after school and work on the computers there. When we finally scraped up enough money to buy a used Commodore 64 in like 1992, I had that hooked up to an old black-and-white TV and taught myself 6502 assembly.
So yeah, I'm sick of people saying "it's someone else's fault that I can't do this." No, it's not. If you're passionate enough about it, you'll *make* it happen.
Slashdot Mirror
Your Usenet post describing the first public version of the Linux kernel compared and contrasted it with GNU, which itself was designed to be very Unix-like. Did you have a "final vision" for Linux in your head in the early days? I mean that as, did you see Linux getting to be a suitable Unix replacement that was still very Unix-like and then stopping, or did you have some sort of vision of a radically different operating system built on top of your kernel?
Obviously, you use Linux every day. What distribution do you use? Do you have a preferred desktop environment or window manager? A preferred text editor?
Australia has corporate tax rates that are in general lower than those in, say, the United States. The US has lower tax rates for corporations with income less than $100,000, but I would very much assume that this studio made more than that. Regulations for this sort of industry are essentially the same around the world as well.
The cost of doing business in Australia is negatively impacted because of major time zone differences from other English-speaking nations, and the significantly higher costs of transportation to/from and telecommunications with what is really a very geographically isolated nation.
How about the 'Hot Dog Stand' theme? That was handy too!
I was mischievous, not evil!
Hah. On the Windows 3.1 systems at my high school I would change the screensaver message to something like "FUCK THA POLICE" or whatever and then use the ATTRIB command to mark WIN.INI as read-only, meaning it was impossible to change the message back using the UI.
Little do they know, they ARE doing the experiment in a simulation.
I have not seen a laptop in close to a decade that didn't work 100% out of the box with Linux. Ubuntu on my laptop actually supported more of the hardware than OEM Windows did without having to find third-party drivers on some website. I remember having to manually configure my sound card with Linux, but I don't remember the last time I did it; it's been at least a decade if not more.
So yeah, the old objections to Linux are, to use an appropriately old term, FUD.
The main thing I'd like to see for 4.0 is a massive simplification of the kernel, removing features that are no longer used anywhere. There's a lot of duplicated functionality in the kernel - two different ways to report hotplug events, three different ways to report ACPI events, several dozen filesystems, some of which don't actually work or even have userland tools that will compile on modern kernels.
So I'd like to see a winnowing of kernel features, down to a saner, all-known-to-work set.
Also, can we please at some point make /proc a bit more sane? Why do I write kernel configuration parameters to files in proc? Why not to sysfs or configfs? /proc, IMHO, should just hold process information. It's confusing.
Blogs are on the internet.
I would argue that there should be some sort of regulation that ensures that phones are interoperable with one another for "phone stuff". That is, if you sell a phone in this country, by law it should be able to make a phone call to every other phone sold in the country. The problem is, what qualifies as "phone stuff" is rapidly expanding.
iMessage is a good example. Apple is trying to leverage its dominant market position to make text messaging something that's iPhone only. Remember the whole debacle with people who had an iPhone and then didn't suddenly not being able to receive text messages from other people who still had iPhones. Apple's solution was broken and only partially effective - and I think at least somewhat intentionally so. Same with FaceTime. You want to talk to your friends with an iPhone? Well, you need an iPhone too!
So yeah, we as a society need to decide what we define as "phone stuff". Having the ability to communicate with every other phone for "phone stuff" is critical from an economic perspective, and eventually will also be so from a safety perspective. Requiring inter-phone communications to be standardized isn't too far-fetched of an idea.
(Requiring the same non-phone-stuff apps to work on different platforms though is stupid.)
This sounds a little insensitive, but, don't be disposable. You're a Windows admin. Great. So are a million other people. If you're a Windows admin who also knows some programming, there are maybe 250,000 people with your skill set. If you add in that you know some Linux, maybe 100,000 people.
What I'm saying is, if you want to be safer than the average employee, don't be average. Enhance your skill set.
You know someone is going to come in and say this is awful because reasons, because it was done under the Obama administration by Eric Holder.
This isn't even my first Slashdot account, my old one dates from 1998 or 1999.
When I first got on Slashdot, there was meaningful, technical discussion. A good number of actual experts in scientific and technical fields were present. Yeah there were always trolls and people racing to have the initial comment, but I feel like the entire tone of Slashdot has changed. You rarely get technical experts on here anymore, the trolls are just as prevalant if not more, and the entire readership has turned a lot to the reactionary right - scientific stores get inundated with "but that's socialism!" comments. There's even a fair amount of junk science in the comments now.
I read Slashdot now solely for the headlines, primarily because I never got the hang of Reddit. The comment section has been basically useless for a while now.
That is all.
There was the botched iOS 8 update that broke phones' data connectivity, and required lots of phones to require reinstallation via iTunes. Then the fix they released was broken and they needed to release a fix for the fix. There might have been a third fix, I can't remember.
Seriously, if my phone isn't capable of doing whatever task I need to do, it means I probably need my full-on laptop anyway. Add in the fact that a tablet either requires wifi or requires cell service but can't make calls and it becomes obvious why the market is behaving this way.
So over my nearly 20 years in IT/CS, I've seen a few:
I worked for a large retailer. We migrated from an old frame-relay leased-line network to a much more capable multihomed IP-over-VPN configuration to connect all of our retail locations around the country back to HQ. This new system worked well. Our CIO retired, and a new one was brought in. CIO Magazine a year or so later had an article about "Satellite Internet, The Future?" Our CIO then "spontaneously" started lobbying to get us to scrap our efficient, inexpensive, high-bandwidth network for a satellite system.
I can't tell you how many projects I saw rewritten in Ruby on Rails just because that was the new hotness, only to be abandoned later when everyone realized that Ruby is awful.
I myself wrote a bunch of stuff in Erlang not because it was the best language but because that was the new hotness.
Two unchanging things I've noticed are:
A lot of time, the new hotness makes common problems go away or common tasks easier, but ends up making more complex things harder. This isn't necessarily a bad thing, but people tend to get stuck in the model of thinking that the new technology has to be used for everything, and they end up shoehorning their complex projects into frameworks that aren't the best choice.
No matter what the new technology is, and no matter how fantastic it is, it's not going to replace C/C++ for systems-level work, and Python and Perl aren't going anywhere. Truly successful technologies have long tails.
I would genuinely appreciate any stories you have to share - not because I don't believe you, but because Slashdot has recently become so "angry white male" that it's depressing.
(I've been on Slashdot since 1998. I remember when article comments were a welcoming forum. Now it's yet another forum where "liberal" on its own is conisdered an insult and labeling someone as such immediately means they are no longer worthy of consideration.)
Heh, right, I mean...I guess I don't understand why one is rotated 180 degrees...It's not that important. :)
Reversed?
Go was developed in large part by Rob Pike who has a long history of concucrrency programming going back to Plan 9 from Bell Labs and earlier.
Some of his more interesting papers about concurrency are:
http://swtch.com/~rsc/thread/n... (The Newsqueak Programming Language)
http://swtch.com/~rsc/thread/n... (Newsqueak Implementation)
https://www.usenix.org/legacy/... (A Concurrent Window System)
You can even see some hints of what was to come in his paper outlining the design of the Blit terminal for Unix:
http://doc.cat-v.org/bell_labs...
I read that as "eat chocolate" even after readreading it twice. I still would've been interested, though, since it's toxic to some mammals.
Greg Egan's stuff wouldn't translate well to the screen, I think. I absolutely love his work (Permutation City is one of my favorite books, and I loved Schild's Ladder, Quarantine, and all of his short fiction). The problem is that there's too heavy of a cognitive science/philosophical bent to them. You'd have to have a character sit down a monologue for a while to get everything out.
Make a Rendezvous With Rama movie, would ya?
Okay, here's the deal. I am passionate about computer science and programming. It's what I do, both for my job, and as my only hobby. I write code for open source projects, and I write code for work, and I design little one-off projects for my own entertainment.
I stayed up all night every summer growing up teaching myself how to code. When I go to the used book store, I go to the section and buy old computer science textbooks talking about esoterica (I'm the only person I know under 45 who knows any APL, for example). My bedtime reading last week was the Oberon System manual that I got off eBay for $5.00.
All this was despite the fact that I grew up in rural Texas and got my ass beaten on a daily basis for being a "geek". The fact that my family was the only non-Christian family in town meant that I couldn't go to the school administration for help; when I tried it turned into a "let's pray for you, son." And yet, I kept doing it because I was passionate about it.
And guess what? If you're that passionate about something, you'll do it regardless of what your peers think. You'll *make* it happen. We didn't have any money growing up, so I'd stay after school and work on the computers there. When we finally scraped up enough money to buy a used Commodore 64 in like 1992, I had that hooked up to an old black-and-white TV and taught myself 6502 assembly.
So yeah, I'm sick of people saying "it's someone else's fault that I can't do this." No, it's not. If you're passionate enough about it, you'll *make* it happen.