^^ What he said, but will also add that all these systemd conspiracy theories about RedHat wanting to make World use systemd seem a bit far fetched. I think that it's more likely, that distros liked enough aspects of systemd to want to incorporate it. All I see in these discussions about Gnome3 and systemd is the same old arguments regurgitated over and over again, by people that probably don't know what they are talking about, or are simply 'confused'. Lennart might well be an asshat--I wouldn't personally know, but if he makes good code that solves problems then fair enough.
I see some comments every now and again, thet seem to give technical credence to systemd, and explain technically why it solves certain problems. I also see comments from people that debunk the half baked theories about why systemd is shite, but what I never see is people debunking *those* posts.
"many eyes make all bugs shallow" is logically correct. Take a company with 200 developers working on closed source. Take an open source project which has *the potential* for all developers in all companies that use that FOSS software to be able to look at the source. Not talking users here but all developers worldwide that incorporate that FOSS project into their work or use it in some fashion. It is a statistically higher probability that a bug *has the potential* to be identified quicker and/or fixed quicker with FOSS than with closed source. ESR's statement still stands.
Now just because there have been bugs that have been around unidentified for decades, does not mean that this is the norm, or that these cases are worse than closed source or an indication that FOSS is wrong in the many eyes theory. You're just trolling.
Btw, "professionals"? Are you serious? You seriously think that FOSS developers are inferior in their competency compared to a developer who works at a company? You don't have a clue. I work in a large dev team for enterprise software (im not a dev). I have the greatest respect for FOSS developers, because with them, it's a pasion, with devs@company it's a pay check.
gotta debunk.
"better choice for Windows users"? Windows 8 is radically different that every other Windows Desktop there has been. Windows 95 was quite different than Windows 3.1 You've got Unity, Gnome, Windows 8, OSX, which are all departures from win95 desktop layouts, and then the rest of the so-called "traditional" Win95-esque desktops. To single Gnome out as something that is so different that it is unusable is a poor argument. By that argument, nobody would upgrade to Win8 because it's too different. But yet all those millions of Windows 7 home users will happily upgrade to Win8 no questions asked, just because it's the newest version. And when even Microsoft changes their own desktop paradigm to something radically different from the Win95-esque, then you've got to ask what you people are smoking when you posit that Win95 is the desktop standard to stick to.
I'm neither a gun owner or American, but your statement just failed a logic test... How do you know that 2.5 million crimes were avoided if they never happened? Are there strictly defined scenarios that have to be met for the police or courts to come to the conclusion that it was actually because the potential victim had a gun and for no other reason, that the crime did not occur? Or are you just quoting some book?
I took the introduction to artificial intelligence course while he was still at Stanford and they tried this MOOC thing there for the first time. I completed it but didnt take any exam. Afterwards, I heard that he was so excited by the whole MOOC thing that he decided to quit his tenure and go to Udacity. I remember thinking at the time that he was being extremely naive in quiting a position at Stanford for something like this, and really got the impression from his video statements that he was being duped, in a way, by Udacity, who likely probably promised him that this was something BIG! Yes, it could have been, but quitting your day job after the first course is a bit drastic.
really, Quake Live is the ultimate fps deathmatch skill game. You can't take any other modern fps storied game, and make it into the competition game that QL is.
CS is the same thing but for tactical combat rather than DM/TeamDM/CTF/duel. That's why CS and Quake Live are till going strong after 16 years. Unless someone takes id tech 5 and releases a new game with heavily updated graphics that also improves even further on weapon/mod balancing,map design and network efficiency in the engine, then QL is going to stay.
Simply put, how long does it take to get something like an Oracle DB up, running and usable on Windows vs Linux? What is the cost of that build, including the licensing and the time it takes to put together? I can image a Linux based server with only the stuff I need significantly faster than I can do the same in Windows Server 2012.
Shame you chose Oracle as your example because it's actually much quicker to install on Windows than Linux/UNIX because you have to faff about creating users/groups and setting kernel parameters and checking you've got all the correct dependency versions. Oracle Linux makes it easier by providing a meta package, but RedHat it's all manual. But your point was still bang on the money. --speaking as an Oracle DBA
I don't upgrade my system just for extra speed or cores, although that's nice with more cores when you transcoded video. For me its more about the surrounding technologies like chipsets, PCI speeds USB speeds especially sata upgrades, because the bottleneck is still the damn disc drive even with ssd.
And every ten years Alzheimer's organisations say that they'll have a cure in ten years time. I see some correlation with that tactic and the general lifespan of Alzheimer's patients.
I've been working as an Oracle DBA in one of the global top 5 IT services companies (and with that said, I only mean that we have a lot of other companies that we host and manage, and im not implying that im a hot shot DBA) for 12 years. Hosting 150 production databases with many diverse setups is a joy to experience. I've worked in both dev and production teams. I've never worked with any other DB, but im a Linux fanboi and would love to get to grips with postgresql as it fits more in line with my ethics than Oracle does.
Long disclaimer aside; I crave comparisons such as this, but every single time I realise that they are all just anecdotes, and every dba on the planet has their own anecdote. It's hard to find comparisons on the Internet from people that are actually qualified to give such comparisons. The closest we can get is face-offs with the likes of Tom Kyte or Jonathan Lewis against similarly experienced MySQL/Postgresql experts. Many Oracle DBA's are also very arrogant, in my opinion, so these anecdotes tend to fly fast.
my experience is that Oracle is extremely expensive, but i've never dealt with paying that money so i've no appreciation of just how bad it can get. Oracle has some technologies like RAC, Data Guard, partitioning and replication, that from what I can gather have no real competition from other vendors, because they are very stable, fast and feature rich in comparison, which I can attest to. Sure, many other DB's have some implementation of these features, but they are not comparable.
Oracle is very configurable, but also very inconsistent in the syntax for configuration, and everything they provide seems to have followed an organic growth method, rather than an overall encompassing design. A lot of technologies are designed by building on how the underlying engine works. For instance, RAC is constrained by the design of the SGA, rightly or wrongly. DG is constrained by the design of redo log generation. From the very tiny exposure i've had to Postgresql, their design philosophy seems to follow the UNIX/Linux philosophy, and I can imagine that if Oracle had followed suit, then some features in oracle may well have been much more efficient and less complex.
I love working with Oracle DB but then again, im a bit of an elitist, so it appeals to me. I do however get the feeling that it could be a hell of a lot less complex and still fulfill the same goals. And I know for a fact that many Oracle customers don't really need that level of complexity. They could have functioned just fine on any other DB motor, especially the ones that don't use licensable features of oracle, but marketing has lead many to believe it's the only choice if your serious about your data.
You conveniently forget that he went to China first and would have stayed in all liklihood if not for the fact that the USA has very strong financial ties and influence there and his security could not be guaranteed. He was probably advised to go to Russia. But don't worry, you're not alone in your forgetfulness... Even John Kerry said the same on TV yesterday
and after RTFA, I could find no mention of the name of "the offical" from Washington anywhere. Almost as if they didn't want to mention their name. Huh. I wonder why? I'd say that was either poor or deceptive reporting from that newspaper.
The only name mentioned in the last paragraph was "State Department spokeswoman Jen Psaki", but it is not certain that she was the same official making the statement.
It's racist to forbid a certain race of people from attending a civilian public conference.
People post on the internet anyway. The information will become free, so unless the US creates a great firewall of USA, they aren't going to stop the information from reaching other countries anyway.
The US, by following through on this threat, would create a precedent for similar actions at other venues, and for other countries to follow suit.
It's a short sighted and immoral suggestion. Which individual suggested this action again?
they're all crap, both in hw and sw. Yes, you can install dd-wrt ét al, but it's still generally crap hardware you're installing it on, and the firmware is only valid for certain hardware.
I bought a Fit-PC 1.0 second hand for 500SEK ($76) and installed pfSense on it which is not hardware specific. Best router i've ever had. I have to use another device for wireless, but there I just re-used my old Linksys WRT610nv2 wireless router as a simple access point. You can get wireless-N access points pretty cheap these days, because the market is all for all-in-one router/switch/AP devices which generally cost $150 or more.
in the same vein of making claims without knowing what im talking about...
I agree, there are lots of skimmers at work, and even if the banks absorb the cost (most of the time), and even if it's only a fraction of what they earn, i would hardly say that it's efficient. Plus, the 'only' loss is virtual currency, which is easy to replace if the bank accepts the loss.
If we're talking about passwords to systems, it's data that is stolen, which can have a more profound impact on the person losing the data. So saying that we could go to PIN codes and use the same system as banks would not be evaluating the risk correctly.
I pay 79 SEK/mnth for Netflix and 99 SEK/mnth for Spotify. I get way more value out of Netflix I feel, because I spend more time on Netflix than I do on Spotify. If they increase by 2 dollars, it brings it up to the same price as Spotify, which I still find reasonable. They probably also figured that Spotify had no reason to charge more for their service as they are similar models.
I think that Gnome3 gets more than it's fair share of criticism by a core group of people, and all I ever seem to see are posts like yours. They invariably go like this:
- Gnome sucks - feature x doesn't work for me ## where feature x is nothing major - feature y doesn't work for me ## where feature y is important, but not neccessarily a core issue - there's a bunch of other features that suck as well, but I don't have time to list them right now
You complain about some valid stuff and maybe some pet gripes, but you write off the *whole desktop* paradigm of Gnome3?
I'd respect your post a lot more if you actually listed your issues, and the reasons why they are issues (the clock? seriously? It's such a big issue for you?), but invariably, personal perspective comes into it, and the logic ends up failing in many cases.
Im a techie, so bit more of a power user than normal, but not a developer with any special use-cases. I think that the *layout and organisation* of Gnome3 is the best ever desktop i've used. At first, I was unsure of the reasons behind certain decisions, but I found a bunch of videos on the Gnome website that explained the reasoning behind it, and after watching them, I realised their logical truth. I wish i could find them again.
Sure there are a bunch of things that i'm not happy about but the general feel of this new desktop paradigm is really really working for me om my stationary PC with 23" monitor.
Happened to me once. I drove all day between sites fixing it equipment in northern England. On the motorway one day I was so tired I feel asleep. I woke up moments later to the sound of traffic cones flying over the front of the car, and realised I was driving on the hard shoulder and that the traffic cones saved my life by waking me up before I drove over the embankment. Scary stuff.
I'm not a mainframe guy and don't know much but from what I've read over the years they're pretty good at being able to handle similar situations. Can they not also handle rollbacks?
You brought up a point I was going to make... they did exactly the same thing, in Paris already, years ago! Was it really '97? I thought that it was permanent then, and did not realise it was only for one day.
Wow, is it true? This is the guy that was all about willingly making it easy for folk to install proprietary drivers for everything to ease adoption of Ubuntu. I remember all the forum discussions about that. Has he finally had a change of heart? RMS is likely having a moment of grim satisfaction right now.
"On the other hand, if you mandate heavyweight regulation in such an industry, you're going to see prices go up significantly, and a lot of useful free-as-in-beer software would probably disappear almost overnight because the people writing it are going to be reluctant to accept engineering-level liability for work they do at charity/PR level prices."
I don't think that's a likely assessment of what could happen if regulation was put in place. Why would free software be in any way bound by regulatory rules? I don't think that would happen, because even most bone-heads would be easily able to grasp that you cannot hold a private person accountable to industry regulations when they release something for free. Fair enough, distribution maintainers that are publicly traded companies that release free software _may_ be affected by regulations, like redhat and canonical, but then they'd just start putting clauses in their distro EULA that the work is from many individuals and they cannot guarantee the quality. It's a bit of a different thing for a single corporation such as Oracle, who holds all the responsibility for a piece of software: then it would be acceptable that they would be bound by regulations.
I don't understand your thinking behind why free software would be affected?
Slashdot Mirror
^^ What he said, but will also add that all these systemd conspiracy theories about RedHat wanting to make World use systemd seem a bit far fetched. I think that it's more likely, that distros liked enough aspects of systemd to want to incorporate it. All I see in these discussions about Gnome3 and systemd is the same old arguments regurgitated over and over again, by people that probably don't know what they are talking about, or are simply 'confused'. Lennart might well be an asshat--I wouldn't personally know, but if he makes good code that solves problems then fair enough.
I see some comments every now and again, thet seem to give technical credence to systemd, and explain technically why it solves certain problems. I also see comments from people that debunk the half baked theories about why systemd is shite, but what I never see is people debunking *those* posts.
I pay $40 for 500/500 in Sweden via Bredband2. I feel sorry for Americans...must be very frustrating.
you're wrong.
"many eyes make all bugs shallow" is logically correct. Take a company with 200 developers working on closed source. Take an open source project which has *the potential* for all developers in all companies that use that FOSS software to be able to look at the source. Not talking users here but all developers worldwide that incorporate that FOSS project into their work or use it in some fashion. It is a statistically higher probability that a bug *has the potential* to be identified quicker and/or fixed quicker with FOSS than with closed source. ESR's statement still stands.
Now just because there have been bugs that have been around unidentified for decades, does not mean that this is the norm, or that these cases are worse than closed source or an indication that FOSS is wrong in the many eyes theory. You're just trolling.
Btw, "professionals"? Are you serious? You seriously think that FOSS developers are inferior in their competency compared to a developer who works at a company? You don't have a clue. I work in a large dev team for enterprise software (im not a dev). I have the greatest respect for FOSS developers, because with them, it's a pasion, with devs@company it's a pay check.
gotta debunk. "better choice for Windows users"? Windows 8 is radically different that every other Windows Desktop there has been. Windows 95 was quite different than Windows 3.1 You've got Unity, Gnome, Windows 8, OSX, which are all departures from win95 desktop layouts, and then the rest of the so-called "traditional" Win95-esque desktops. To single Gnome out as something that is so different that it is unusable is a poor argument. By that argument, nobody would upgrade to Win8 because it's too different. But yet all those millions of Windows 7 home users will happily upgrade to Win8 no questions asked, just because it's the newest version. And when even Microsoft changes their own desktop paradigm to something radically different from the Win95-esque, then you've got to ask what you people are smoking when you posit that Win95 is the desktop standard to stick to.
I'm neither a gun owner or American, but your statement just failed a logic test... How do you know that 2.5 million crimes were avoided if they never happened? Are there strictly defined scenarios that have to be met for the police or courts to come to the conclusion that it was actually because the potential victim had a gun and for no other reason, that the crime did not occur? Or are you just quoting some book?
I took the introduction to artificial intelligence course while he was still at Stanford and they tried this MOOC thing there for the first time. I completed it but didnt take any exam. Afterwards, I heard that he was so excited by the whole MOOC thing that he decided to quit his tenure and go to Udacity. I remember thinking at the time that he was being extremely naive in quiting a position at Stanford for something like this, and really got the impression from his video statements that he was being duped, in a way, by Udacity, who likely probably promised him that this was something BIG! Yes, it could have been, but quitting your day job after the first course is a bit drastic.
really, Quake Live is the ultimate fps deathmatch skill game. You can't take any other modern fps storied game, and make it into the competition game that QL is. CS is the same thing but for tactical combat rather than DM/TeamDM/CTF/duel. That's why CS and Quake Live are till going strong after 16 years. Unless someone takes id tech 5 and releases a new game with heavily updated graphics that also improves even further on weapon/mod balancing,map design and network efficiency in the engine, then QL is going to stay.
Simply put, how long does it take to get something like an Oracle DB up, running and usable on Windows vs Linux? What is the cost of that build, including the licensing and the time it takes to put together? I can image a Linux based server with only the stuff I need significantly faster than I can do the same in Windows Server 2012.
Shame you chose Oracle as your example because it's actually much quicker to install on Windows than Linux/UNIX because you have to faff about creating users/groups and setting kernel parameters and checking you've got all the correct dependency versions. Oracle Linux makes it easier by providing a meta package, but RedHat it's all manual. But your point was still bang on the money. --speaking as an Oracle DBA
I don't upgrade my system just for extra speed or cores, although that's nice with more cores when you transcoded video. For me its more about the surrounding technologies like chipsets, PCI speeds USB speeds especially sata upgrades, because the bottleneck is still the damn disc drive even with ssd.
God damn beta and god damn tablet software keyboards
And every ten years Alzheimer's organisations say that they'll have a cure in ten years time. I see some correlation with that tactic and the general lifespan of Alzheimer's patients.
I've been working as an Oracle DBA in one of the global top 5 IT services companies (and with that said, I only mean that we have a lot of other companies that we host and manage, and im not implying that im a hot shot DBA) for 12 years. Hosting 150 production databases with many diverse setups is a joy to experience. I've worked in both dev and production teams. I've never worked with any other DB, but im a Linux fanboi and would love to get to grips with postgresql as it fits more in line with my ethics than Oracle does.
Long disclaimer aside; I crave comparisons such as this, but every single time I realise that they are all just anecdotes, and every dba on the planet has their own anecdote. It's hard to find comparisons on the Internet from people that are actually qualified to give such comparisons. The closest we can get is face-offs with the likes of Tom Kyte or Jonathan Lewis against similarly experienced MySQL/Postgresql experts. Many Oracle DBA's are also very arrogant, in my opinion, so these anecdotes tend to fly fast.
my experience is that Oracle is extremely expensive, but i've never dealt with paying that money so i've no appreciation of just how bad it can get. Oracle has some technologies like RAC, Data Guard, partitioning and replication, that from what I can gather have no real competition from other vendors, because they are very stable, fast and feature rich in comparison, which I can attest to. Sure, many other DB's have some implementation of these features, but they are not comparable.
Oracle is very configurable, but also very inconsistent in the syntax for configuration, and everything they provide seems to have followed an organic growth method, rather than an overall encompassing design. A lot of technologies are designed by building on how the underlying engine works. For instance, RAC is constrained by the design of the SGA, rightly or wrongly. DG is constrained by the design of redo log generation. From the very tiny exposure i've had to Postgresql, their design philosophy seems to follow the UNIX/Linux philosophy, and I can imagine that if Oracle had followed suit, then some features in oracle may well have been much more efficient and less complex.
I love working with Oracle DB but then again, im a bit of an elitist, so it appeals to me. I do however get the feeling that it could be a hell of a lot less complex and still fulfill the same goals. And I know for a fact that many Oracle customers don't really need that level of complexity. They could have functioned just fine on any other DB motor, especially the ones that don't use licensable features of oracle, but marketing has lead many to believe it's the only choice if your serious about your data.
You conveniently forget that he went to China first and would have stayed in all liklihood if not for the fact that the USA has very strong financial ties and influence there and his security could not be guaranteed. He was probably advised to go to Russia. But don't worry, you're not alone in your forgetfulness... Even John Kerry said the same on TV yesterday
and after RTFA, I could find no mention of the name of "the offical" from Washington anywhere. Almost as if they didn't want to mention their name. Huh. I wonder why? I'd say that was either poor or deceptive reporting from that newspaper.
The only name mentioned in the last paragraph was "State Department spokeswoman Jen Psaki", but it is not certain that she was the same official making the statement.
It's racist to forbid a certain race of people from attending a civilian public conference.
People post on the internet anyway. The information will become free, so unless the US creates a great firewall of USA, they aren't going to stop the information from reaching other countries anyway.
The US, by following through on this threat, would create a precedent for similar actions at other venues, and for other countries to follow suit.
It's a short sighted and immoral suggestion. Which individual suggested this action again?
they're all crap, both in hw and sw. Yes, you can install dd-wrt ét al, but it's still generally crap hardware you're installing it on, and the firmware is only valid for certain hardware.
I bought a Fit-PC 1.0 second hand for 500SEK ($76) and installed pfSense on it which is not hardware specific. Best router i've ever had. I have to use another device for wireless, but there I just re-used my old Linksys WRT610nv2 wireless router as a simple access point. You can get wireless-N access points pretty cheap these days, because the market is all for all-in-one router/switch/AP devices which generally cost $150 or more.
in the same vein of making claims without knowing what im talking about...
I agree, there are lots of skimmers at work, and even if the banks absorb the cost (most of the time), and even if it's only a fraction of what they earn, i would hardly say that it's efficient. Plus, the 'only' loss is virtual currency, which is easy to replace if the bank accepts the loss.
If we're talking about passwords to systems, it's data that is stolen, which can have a more profound impact on the person losing the data. So saying that we could go to PIN codes and use the same system as banks would not be evaluating the risk correctly.
I pay 79 SEK/mnth for Netflix and 99 SEK/mnth for Spotify. I get way more value out of Netflix I feel, because I spend more time on Netflix than I do on Spotify. If they increase by 2 dollars, it brings it up to the same price as Spotify, which I still find reasonable. They probably also figured that Spotify had no reason to charge more for their service as they are similar models.
You've got to have balls to come in here as a member of the gnome team. I'll give you that. Respect.
I think that Gnome3 gets more than it's fair share of criticism by a core group of people, and all I ever seem to see are posts like yours. They invariably go like this:
- Gnome sucks
- feature x doesn't work for me ## where feature x is nothing major
- feature y doesn't work for me ## where feature y is important, but not neccessarily a core issue
- there's a bunch of other features that suck as well, but I don't have time to list them right now
You complain about some valid stuff and maybe some pet gripes, but you write off the *whole desktop* paradigm of Gnome3?
I'd respect your post a lot more if you actually listed your issues, and the reasons why they are issues (the clock? seriously? It's such a big issue for you?), but invariably, personal perspective comes into it, and the logic ends up failing in many cases.
Im a techie, so bit more of a power user than normal, but not a developer with any special use-cases. I think that the *layout and organisation* of Gnome3 is the best ever desktop i've used. At first, I was unsure of the reasons behind certain decisions, but I found a bunch of videos on the Gnome website that explained the reasoning behind it, and after watching them, I realised their logical truth. I wish i could find them again.
Sure there are a bunch of things that i'm not happy about but the general feel of this new desktop paradigm is really really working for me om my stationary PC with 23" monitor.
Happened to me once. I drove all day between sites fixing it equipment in northern England. On the motorway one day I was so tired I feel asleep. I woke up moments later to the sound of traffic cones flying over the front of the car, and realised I was driving on the hard shoulder and that the traffic cones saved my life by waking me up before I drove over the embankment. Scary stuff.
I'm not a mainframe guy and don't know much but from what I've read over the years they're pretty good at being able to handle similar situations. Can they not also handle rollbacks?
You brought up a point I was going to make... they did exactly the same thing, in Paris already, years ago! Was it really '97? I thought that it was permanent then, and did not realise it was only for one day.
Wow, is it true? This is the guy that was all about willingly making it easy for folk to install proprietary drivers for everything to ease adoption of Ubuntu. I remember all the forum discussions about that. Has he finally had a change of heart? RMS is likely having a moment of grim satisfaction right now.
"On the other hand, if you mandate heavyweight regulation in such an industry, you're going to see prices go up significantly, and a lot of useful free-as-in-beer software would probably disappear almost overnight because the people writing it are going to be reluctant to accept engineering-level liability for work they do at charity/PR level prices."
I don't think that's a likely assessment of what could happen if regulation was put in place. Why would free software be in any way bound by regulatory rules? I don't think that would happen, because even most bone-heads would be easily able to grasp that you cannot hold a private person accountable to industry regulations when they release something for free. Fair enough, distribution maintainers that are publicly traded companies that release free software _may_ be affected by regulations, like redhat and canonical, but then they'd just start putting clauses in their distro EULA that the work is from many individuals and they cannot guarantee the quality. It's a bit of a different thing for a single corporation such as Oracle, who holds all the responsibility for a piece of software: then it would be acceptable that they would be bound by regulations.
I don't understand your thinking behind why free software would be affected?