Lennart is actually very clear on the technical reasons why he chose to create systemd even if Mark wants to remain ignorant of them.
Those are not "technical" reasons, those are personal.
Basicly, To surmise for the non-developers, Lennart says that he doesn't like programming events. That's pretty much all he says against the Upstart.
P.S. OK, I'm omitting the stupid part where he complains that "but upstart starts everything!", apparently displaying lack of knowledge what the SysV init scripts do (for which Upstart is the replacement).
Many Apple users now have to use the official legal lightning cables included with their IPhone
Well, you don't say.
It's still beyond me.
I frankly have expected Apple to fork/extend/etc the Micro-USB to kill three/more birds with one stone. And I would have most likely applauded the effort.
But they decides to anchor their users with the stone instead.
Jobs gone. There is nobody to guide the engineers.... There is nobody to fend off the MBAs....
That said for the portable device on the other end to recognise charging mode it also needs to see some fixed resistance, which would need to be build into the far end plug too.
Samsung's charger for the Galaxies simply shorts the data pins. (No, not the cable. The charger.) They do it as a way to recognize that it is a charger connected and allow drawing more power.
I'm not sure how it is in the new kernels, but in the past it was basically one flag - one bit - to tell whether the device should contribute to random pool or not. What I'm saying, it is not "ridiculously impractical," it is very highly likely a single line of code.
And since we are in the embedded context, then it is also not particularly hard requirements, since for embedded systems kernel is routinely patched and custom built anyway (for hardware quirks, for extra non-standard hardware, support for specialized software, etc).
"Larger existing libraries" usually come with their own management software. If your private library has several thousands books, then it is not "large", it is small. "Large" libraries are in tens thousand and more.
Import into Calibre has only the effects: it copies the book into Calibre library (or any library you have configures/selected), it extracts the cover and it extracts the meta-data into OPF file. If you do not have library management software, the cover and meta-data extraction are the steps you need anyway and those are the steps which take the most time.
Copying is a feature too. I also have an enormous external library of books - but due to its size it is on the external hard-drive. I connect the drive, pick the books, import them into Calibre and disconnect the drive. Now I can send the books to my Kindle for reading. Huge bonus: every time I start Calibre, I'm not overwhelmed by the sheer amount of the books in the library; I see only the books which I want to see.
I don't like driving during my commute, however I do like driving for fun. Unfortunately driving for fun is beyond my economic means these days. Gas is 6 or 7 times what it was when I was a teenager.
But even if economically feasible, for those of us who live in cities, "driving for fun" means: 30+ minutes of city traffic; 30+ minutes to get anywhere interesting; ah, the joy of driving!!!; spend 60+ minutes getting back home.
I promised myself that I would buy a car again, only if it can autonomously drive me from A to B to A, aka the commute to/from office.
RTFA is shallow on details, but it sounds very much like the "phantom" self-mutating MS-DOS viruses from the later 90. (The ones which forced anti-virus vendors to effectively implement a debugger/a interpreter for the code to detect the malicious code not by the look but by the actual work it does.)
Otherwise, the main problems with such intrusive obfuscations is that code is never 100% bug free and bugs sometimes trigger undefined/unspecified behaviors. It might work before obfuscation - but would break (in unexplainable, obfuscated fashion) afterwards. (Ditto after system updates.)
A lot of the "I.T. won't buy me a pony" bullshit is because a company expects managers to decide if the junior staff get a pony or not, and when they don't get one they decide to make an end run around their management and try to get it from I.T.
That is different where I'm currently working, because departments foot the bill, not I.T. Effectively, it is something like: "department pays for a race horse but I.T. keeps it, and gives you their old pony instead." And their argument is "pony's fully compatible, it also has four hooves!"
I'm not sure exactly where you've worked, but you're somewhat shrieky about the whole thing.
In your experience. I've been with the same organization for fourteen years, and IT can be a force multiplier.
Three out of four employments I have had in the last 10 years, IT departments were to some extent as I have described. One company was small enough to be perfectly OK (we had one admin, who was really helpful guy). The companies I work for are generally pretty normal software vendors and/or hardware manufacturers.
Also, the fact that your experience as admin is totally different, doesn't really mean anything: I'm pretty sure our admins are also very positive about the work they do. Precisely because they are ignorant of what is going on in the rest of the company. (Like few month ago: two weeks of infrastructure upgrades while all were trying to work on a launch of crucial project. Everyone was assured that servers and services would run interrupted. That was true, we were not lied to. But the fact that we couldn't access the servers because the network was down most of the time... Well, who gives a.)
The first thing I do with new hires is make sure they know exactly what the technology team's mission is (which is supporting the organizational mission and its objectives).
Oh. That definitely puts you and your organization into a different class. What I can only envy.
Me personally? I don't like having the full responsibility for something not working, and the responsibility for making it work, when I was not included AT ALL in any phase of the project.
That is normal for pretty much all businesses.
I work in support now and often end-up being responsible for stupid, short-sighted decisions made during planning and deployment.
But that's what I'm paid for to do, and thus can't complain. I'm not sure what makes admins think differently.
If I am to be held responsible for the entire IT infrastructure than you bloody well better bet I'm going to act like I own it.
And that is the crux of the problem: why the hell IT is blankly made responsible for everything? Because of the stupid idea: "let's spare time of employees dealing with the issue and put dedicated people on it." With the time the gap's between IT and employees widens enough where both parties simply can't talk to each other, because they do not have any sort of common language, nor aware of the problems. It doesn't help when IT's buying decisions are based on how easy something to administer, not how easy it is to use by the users. Neither IT's "one size fits all" approach (aka "harmonization, standardization and unification.") IOW, it starts with the "good" idea of outsourcing problems to another department, and ends with the problems of dealing with the said department.
From my experience, the best working solution is to go away with IT department and the silly notion that the IT should be centralized. If company is large enough to theoretically have an IT department, then certainly it is too large to have it, because "one size NEVER fits all." Put admins like normal employees to the structure, reporting to department managers. If something company-wide needs to be implemented, then admins from different departments can meet, present requirements of the departments and agree on the solution and the scope of the solution. Or, probably, agree that it can't be made company-wide because requirements are contradictory. Otherwise, admins sit with the users and are aware immediately of the user's problems - and vice versa: users are aware of the administrative problems. I have seen that kind of structure implemented only once, in a mid-sized company, and it worked pretty well (from admin's own words).
The worst characteristics of Sysadmins tend to emerge when the organization treats them badly.
I yet to have an employer who treats sysadmins badly.
And if you treat admins well or very well, the IT becomes a money sink hole, with admins generally caring about their own problems. They feel like they own the whole IT infrastructure and can do with it whatever they want, not giving a single fuck about people who do the actual work of the company on that infrastructure. (Hey! ITs often don't even know what company actually does!!) Oh, and when some problem gets really escalated by the employees, IT often makes sure that the meeting is held only with the managers, who can't refute their "arguments." After all, the stereotypical manager can be always convinced with the unbeatable irrefutable argument "if we do that, the Exchange server might break!"
IMO, the stereotypes appear not because admins go bad in toxic environments. But because IT often degenerates into a self-centered parallel universe of its own, what eventually de-evolves into the toxic environment.
Mac OS X, since the early version has excellent support for keyboard shortcuts. (Albeit tricky to configure.) And applications are always installed in one single location on Mac OS X - unlike Windows where some are in Program Files, some in Program Files x64, some in Windows, some in system32, etc. And applications on Mac OS are represented with a single user-friendly icon - not a folder with pile of subfolders where you still have to hunt for the proper executable. Bonus: the Dock (now also in Windows since 7) was always there to quickly access often used applications.
And finally, since the introduction of the whole-OS-and-hard-drive search function, it is a matter of pressing Cmd+Space, typing application name and pressing Enter. (Bonus: pressing Cmd+Space and searching doesn't steal focus, pressing Win to access Win7 menu's search does still the focus from active application. (Win8 - it's not only steals focus, it's switching whole desktop to different UI mode.) Some Windows applications have problems properly restoring focus where user left it before Alt+Tab or the focus steal.)
All in all, Mac OS was made from ground up to live without the Start button. And as such, many functions are provided to access applications and whatnot quickly. On Windows, the MSFT never really bothered to figure out how users actually use the frigging Windows. Metro is not about improvements for the user - it is about sneaking Windows into the tablet market.
P.S. Do not get me wrong. I'm not huge fan of Mac OS X interface. For example, its nested switching between windows (first switch between applications, then switch between windows; and no, Expose is not the answer) is a horrible cludge.
Maybe you can answer this since the website doesn't. What is the difference between aptosid and just running sid?
Many things were listed before, but the biggest ones were not:
1. You can't install Debian Sid, since there is no installer for it. You need to install normall Debian, and then add the sid repos and dist-upgrade. Can be painful and very error-prone (see #2).
2. As much as people like to say that Aptosid (formerly Sidux) is pure Sid, it is not. The Aptosid maintainers/Debian developers try to hold off known to be broken version of packages.
And another little gem - OpenNTPD. Used on several embedded systems. Works like a charm, unlike the whimsical ntpd, which often simply refuses to do its job.
I can't recall the year/etc, but your fancy chart is missing literally the whole Win NT 3.x branch. I can remember seeing the Win NT 3.1 box (I had also impression at the time that there was also NT 3.0). And the NT 3.50. And the best of them all - the NT 3.51.
Strangely enough I can't recall any mention of the 3.46.
It looks like Mozilla are finally going to remove the much hated blink tag
I do not hate the blink tag. I hate the web developers.
Otherwise, I think the blink tag spurred the whole generation of web developers. Just look at all the so-called "Web 2.0" crap. Whether it is <blink> or jQuery's animations for every however tiny P.O.S., the end result is the same: unusable mess.
MEncoder definitely capable of extracting part of the movie(*) and combining several ones together. (*)Though IIRC at least in the past MEncoder insisted on frame numbers and wasn't accepting simple times. (I had very little luck with the MEncoder since it often screwed up the A/V sync. But apparently it works for many, since literally all video reencoders for iPhone/PSP/PS3/etc are based on it.)
MKVMerge can't reencode and as such is more limited. But often is sufficient and produces very good results. It definitely can extract part of input. Joining several movies together I have never tried, but googling says it is also possible. (Nice thing about MKVMerge is that it has GUI and CLI. GUI, after clicking all desired options and whatnot, provides you with the command line to be used to start the actual CLI mkvmerge. You can take the command and tweak it to your heart content. I have used that to remux batch of movies with different sound and subtitles, by simply replacing input/output file names in the command line.)
Overall, what you ask is literally impractical: extracting creates a copy of a movie (large input = large output), combining/appending is potentially reencoding (output must be encoded homogeneously, while input is not guaranteed to be homogeneous) and thus very slow. You might wait an hour for the command to finish, only to find that you have cut too much or too little. That is why the video editors are GUI: you select the inputs, you tell it what to you want from it and then you can preview the end result, without waiting hours for the actual rendering.
P.S. Long in the past I have also used the Transcode. Worked pretty well, though is limited to AVI and MPEG2. But it does seem to be abandoned now.
Because one never knows what motivates the "masterminds" up in the HQ - and they might really close the YouTube if they feel like it. Especially after they would 100% integrate it with the G+ and drive users off.
Slashdot Mirror
Lennart is actually very clear on the technical reasons why he chose to create systemd even if Mark wants to remain ignorant of them.
Those are not "technical" reasons, those are personal.
Basicly, To surmise for the non-developers, Lennart says that he doesn't like programming events. That's pretty much all he says against the Upstart.
P.S. OK, I'm omitting the stupid part where he complains that "but upstart starts everything!", apparently displaying lack of knowledge what the SysV init scripts do (for which Upstart is the replacement).
Many Apple users now have to use the official legal lightning cables included with their IPhone
Well, you don't say.
It's still beyond me.
I frankly have expected Apple to fork/extend/etc the Micro-USB to kill three/more birds with one stone. And I would have most likely applauded the effort.
But they decides to anchor their users with the stone instead.
Jobs gone. There is nobody to guide the engineers.... There is nobody to fend off the MBAs....
That said for the portable device on the other end to recognise charging mode it also needs to see some fixed resistance, which would need to be build into the far end plug too.
Samsung's charger for the Galaxies simply shorts the data pins. (No, not the cable. The charger.) They do it as a way to recognize that it is a charger connected and allow drawing more power.
I'm not sure how it is in the new kernels, but in the past it was basically one flag - one bit - to tell whether the device should contribute to random pool or not. What I'm saying, it is not "ridiculously impractical," it is very highly likely a single line of code.
And since we are in the embedded context, then it is also not particularly hard requirements, since for embedded systems kernel is routinely patched and custom built anyway (for hardware quirks, for extra non-standard hardware, support for specialized software, etc).
It's a common trope in USA that most poor people are poor because they're lazy or just inherently bad with money.
FTFY.
Otherwise, I have seen plenty of rich people who were also pretty bad with money.
"Larger existing libraries" usually come with their own management software. If your private library has several thousands books, then it is not "large", it is small. "Large" libraries are in tens thousand and more.
Import into Calibre has only the effects: it copies the book into Calibre library (or any library you have configures/selected), it extracts the cover and it extracts the meta-data into OPF file. If you do not have library management software, the cover and meta-data extraction are the steps you need anyway and those are the steps which take the most time.
Copying is a feature too. I also have an enormous external library of books - but due to its size it is on the external hard-drive. I connect the drive, pick the books, import them into Calibre and disconnect the drive. Now I can send the books to my Kindle for reading. Huge bonus: every time I start Calibre, I'm not overwhelmed by the sheer amount of the books in the library; I see only the books which I want to see.
In the right-bottom corner, uncheck all the buttons except the right-most one (with the icon of a tiles).
Here you go - a modern interface! Wasn't that hard to find too.
Check out the toggle-buttons in right-bottom corner.
Check the "Cover Grid" and uncheck the "Cover Browser", "Tag Browser" and "Book Details". Looks better now?
The last touch: do not use the hideous right-click pop-up menu - but learn to use the buttons on the top. All the functions are there.
I don't like driving during my commute, however I do like driving for fun. Unfortunately driving for fun is beyond my economic means these days. Gas is 6 or 7 times what it was when I was a teenager.
But even if economically feasible, for those of us who live in cities, "driving for fun" means: 30+ minutes of city traffic; 30+ minutes to get anywhere interesting; ah, the joy of driving!!!; spend 60+ minutes getting back home.
I promised myself that I would buy a car again, only if it can autonomously drive me from A to B to A, aka the commute to/from office.
Maybe there's a 3rd party add-on?
Yes.
There are more than one. Search for FlashBlock or AdBlock.
Works for me.
RTFA is shallow on details, but it sounds very much like the "phantom" self-mutating MS-DOS viruses from the later 90. (The ones which forced anti-virus vendors to effectively implement a debugger/a interpreter for the code to detect the malicious code not by the look but by the actual work it does.)
Otherwise, the main problems with such intrusive obfuscations is that code is never 100% bug free and bugs sometimes trigger undefined/unspecified behaviors. It might work before obfuscation - but would break (in unexplainable, obfuscated fashion) afterwards. (Ditto after system updates.)
But Firefox OS is implemented by the very same people.
IOW, expect the same attention to details and customer-friendliness.
A lot of the "I.T. won't buy me a pony" bullshit is because a company expects managers to decide if the junior staff get a pony or not, and when they don't get one they decide to make an end run around their management and try to get it from I.T.
That is different where I'm currently working, because departments foot the bill, not I.T. Effectively, it is something like: "department pays for a race horse but I.T. keeps it, and gives you their old pony instead." And their argument is "pony's fully compatible, it also has four hooves!"
I'm not sure exactly where you've worked, but you're somewhat shrieky about the whole thing.
In your experience. I've been with the same organization for fourteen years, and IT can be a force multiplier.
Three out of four employments I have had in the last 10 years, IT departments were to some extent as I have described. One company was small enough to be perfectly OK (we had one admin, who was really helpful guy). The companies I work for are generally pretty normal software vendors and/or hardware manufacturers.
Also, the fact that your experience as admin is totally different, doesn't really mean anything: I'm pretty sure our admins are also very positive about the work they do. Precisely because they are ignorant of what is going on in the rest of the company. (Like few month ago: two weeks of infrastructure upgrades while all were trying to work on a launch of crucial project. Everyone was assured that servers and services would run interrupted. That was true, we were not lied to. But the fact that we couldn't access the servers because the network was down most of the time... Well, who gives a.)
The first thing I do with new hires is make sure they know exactly what the technology team's mission is (which is supporting the organizational mission and its objectives).
Oh. That definitely puts you and your organization into a different class. What I can only envy.
Me personally? I don't like having the full responsibility for something not working, and the responsibility for making it work, when I was not included AT ALL in any phase of the project.
That is normal for pretty much all businesses.
I work in support now and often end-up being responsible for stupid, short-sighted decisions made during planning and deployment.
But that's what I'm paid for to do, and thus can't complain. I'm not sure what makes admins think differently.
If I am to be held responsible for the entire IT infrastructure than you bloody well better bet I'm going to act like I own it.
And that is the crux of the problem: why the hell IT is blankly made responsible for everything? Because of the stupid idea: "let's spare time of employees dealing with the issue and put dedicated people on it." With the time the gap's between IT and employees widens enough where both parties simply can't talk to each other, because they do not have any sort of common language, nor aware of the problems. It doesn't help when IT's buying decisions are based on how easy something to administer, not how easy it is to use by the users. Neither IT's "one size fits all" approach (aka "harmonization, standardization and unification.") IOW, it starts with the "good" idea of outsourcing problems to another department, and ends with the problems of dealing with the said department.
From my experience, the best working solution is to go away with IT department and the silly notion that the IT should be centralized. If company is large enough to theoretically have an IT department, then certainly it is too large to have it, because "one size NEVER fits all." Put admins like normal employees to the structure, reporting to department managers. If something company-wide needs to be implemented, then admins from different departments can meet, present requirements of the departments and agree on the solution and the scope of the solution. Or, probably, agree that it can't be made company-wide because requirements are contradictory. Otherwise, admins sit with the users and are aware immediately of the user's problems - and vice versa: users are aware of the administrative problems. I have seen that kind of structure implemented only once, in a mid-sized company, and it worked pretty well (from admin's own words).
The worst characteristics of Sysadmins tend to emerge when the organization treats them badly.
I yet to have an employer who treats sysadmins badly.
And if you treat admins well or very well, the IT becomes a money sink hole, with admins generally caring about their own problems. They feel like they own the whole IT infrastructure and can do with it whatever they want, not giving a single fuck about people who do the actual work of the company on that infrastructure. (Hey! ITs often don't even know what company actually does!!) Oh, and when some problem gets really escalated by the employees, IT often makes sure that the meeting is held only with the managers, who can't refute their "arguments." After all, the stereotypical manager can be always convinced with the unbeatable irrefutable argument "if we do that, the Exchange server might break!"
IMO, the stereotypes appear not because admins go bad in toxic environments. But because IT often degenerates into a self-centered parallel universe of its own, what eventually de-evolves into the toxic environment.
... but only for a short time, before user faints from the toxic smoke.
It's ok that OS X has never had a Start menu.
Mac OS X, since the early version has excellent support for keyboard shortcuts. (Albeit tricky to configure.) And applications are always installed in one single location on Mac OS X - unlike Windows where some are in Program Files, some in Program Files x64, some in Windows, some in system32, etc. And applications on Mac OS are represented with a single user-friendly icon - not a folder with pile of subfolders where you still have to hunt for the proper executable. Bonus: the Dock (now also in Windows since 7) was always there to quickly access often used applications.
And finally, since the introduction of the whole-OS-and-hard-drive search function, it is a matter of pressing Cmd+Space, typing application name and pressing Enter. (Bonus: pressing Cmd+Space and searching doesn't steal focus, pressing Win to access Win7 menu's search does still the focus from active application. (Win8 - it's not only steals focus, it's switching whole desktop to different UI mode.) Some Windows applications have problems properly restoring focus where user left it before Alt+Tab or the focus steal.)
All in all, Mac OS was made from ground up to live without the Start button. And as such, many functions are provided to access applications and whatnot quickly. On Windows, the MSFT never really bothered to figure out how users actually use the frigging Windows. Metro is not about improvements for the user - it is about sneaking Windows into the tablet market.
P.S. Do not get me wrong. I'm not huge fan of Mac OS X interface. For example, its nested switching between windows (first switch between applications, then switch between windows; and no, Expose is not the answer) is a horrible cludge.
Maybe you can answer this since the website doesn't. What is the difference between aptosid and just running sid?
Many things were listed before, but the biggest ones were not:
1. You can't install Debian Sid, since there is no installer for it. You need to install normall Debian, and then add the sid repos and dist-upgrade. Can be painful and very error-prone (see #2).
2. As much as people like to say that Aptosid (formerly Sidux) is pure Sid, it is not. The Aptosid maintainers/Debian developers try to hold off known to be broken version of packages.
And another little gem - OpenNTPD. Used on several embedded systems. Works like a charm, unlike the whimsical ntpd, which often simply refuses to do its job.
I can't recall the year/etc, but your fancy chart is missing literally the whole Win NT 3.x branch. I can remember seeing the Win NT 3.1 box (I had also impression at the time that there was also NT 3.0). And the NT 3.50. And the best of them all - the NT 3.51.
Strangely enough I can't recall any mention of the 3.46.
It looks like Mozilla are finally going to remove the much hated blink tag
I do not hate the blink tag. I hate the web developers.
Otherwise, I think the blink tag spurred the whole generation of web developers. Just look at all the so-called "Web 2.0" crap. Whether it is <blink> or jQuery's animations for every however tiny P.O.S., the end result is the same: unusable mess.
You should check both MKVMerge and MEncoder.
MEncoder definitely capable of extracting part of the movie(*) and combining several ones together. (*)Though IIRC at least in the past MEncoder insisted on frame numbers and wasn't accepting simple times. (I had very little luck with the MEncoder since it often screwed up the A/V sync. But apparently it works for many, since literally all video reencoders for iPhone/PSP/PS3/etc are based on it.)
MKVMerge can't reencode and as such is more limited. But often is sufficient and produces very good results. It definitely can extract part of input. Joining several movies together I have never tried, but googling says it is also possible. (Nice thing about MKVMerge is that it has GUI and CLI. GUI, after clicking all desired options and whatnot, provides you with the command line to be used to start the actual CLI mkvmerge. You can take the command and tweak it to your heart content. I have used that to remux batch of movies with different sound and subtitles, by simply replacing input/output file names in the command line.)
Overall, what you ask is literally impractical: extracting creates a copy of a movie (large input = large output), combining/appending is potentially reencoding (output must be encoded homogeneously, while input is not guaranteed to be homogeneous) and thus very slow. You might wait an hour for the command to finish, only to find that you have cut too much or too little. That is why the video editors are GUI: you select the inputs, you tell it what to you want from it and then you can preview the end result, without waiting hours for the actual rendering.
P.S. Long in the past I have also used the Transcode. Worked pretty well, though is limited to AVI and MPEG2. But it does seem to be abandoned now.
motion/visual selection + g? in Vim.
Pretty lame of Google to joke like that.
Because one never knows what motivates the "masterminds" up in the HQ - and they might really close the YouTube if they feel like it. Especially after they would 100% integrate it with the G+ and drive users off.