Re:Basic misunderstandings and self-contradictions
on
Linux Sucks (Video)
·
· Score: 1
The point is we need people like him to remind us that certain things suck and need to be replaced (cough, X11, cough) otherwise we ae stuck with old badly architected technology for decades.
It's hard to find somebody that says X11 doesn't suck. I am definitely not that person.
My point was that he says forking sucks, he gave an example where (unbeknownst to him?) forking was certainly the best option, then he went on to talk about how forked Linux distributions have made the world a better place. He seems to conclude that forking is great and that he "loves" it.
(Also, I misspelled his name. Sorry, Bryan. I guess my post had room for improvement... meaning it sucked.)
Basic misunderstandings and self-contradictions
on
Linux Sucks (Video)
·
· Score: 4, Interesting
OpenSSL doesn't listen to bug reports. They don't even accept offered patches to known bugs. It's this spirit of non-cooperation that caused the forking into LibreSSL. See the 30 day prospectus (/. coverage) from the LibreSSL project lead, which details all of the problems. Brian even says forking is ultimately a benefit, and that he "loves that they're doing it."
It seems to be that his definition of "sucks" is "has room for improvement"... Everything has room for improvement, so apparently everything sucks.
But part of me wonders if I'm missing the point, if they're so intent on breaking it then might I as well just move browsers now? If I'm having to rely on addons to make a browser work then am I not just sat precariously one step away from Mozilla deciding that addon is unacceptable and cancelling it anyway?
It appears the FF devs have forgotten that their main advantage over Chrome is addons. I have so many addons, with icons to control them in the status bar (addons bar) that the new FF gave me about an inch of locationbar to see URLs. Thanks guys. I reverted this by using the dev version of Status-4-Evar. The GP's mention of Classic Theme Restorer is interesting, but I worry about its compatibility with Tab Mix Plus and other addons, as well as to your point of perhaps trying the new look & feel.
The core tenant behind the increasingly popular paleo diet is that food has been over-domesticated, favoring things like size, portability, and crop yield rather than health. Taste is often also low on the priority list (though higher than health). Wild plants like dandelion greens and ramps are significantly healthier than our domesticated cabbages for example.
The same goes for meat. Wild game meat is far healthier than meat from a factory farm. It's often tastier as well, though the farmed stuff tends to be fattier (and fat equals flavor). I'd love to try the meat of an ancestor of the cow that pre-dates its domestication. (It should also be eating and excersizing similar to the way it would in the wild rather than eating corn and living in tight quarters.)
Times have changed - when I did my computer science degree, most of the students were at the geeky end of the spectrum and were there because that's what they were really into. Compare to the present-day cross section of computer science students: most of them are there because computers are seen as a good career. The extra-curricular interest is giving way to people who just want a job.
I disagree. People like you and me merely congregated together and ignored the others. (Also, you went to school in Wales. Different world.) My above statement was about "my most IT-savvy freshman colleagues," which is to say under a dozen total (and I was friends with all of them). I'd say about 75% of my freshman peers in CS declared the major for its salaries and/or a passion for video games. I imagine today's breakdown is roughly the same, more due to the fact that most freshmen are blank slates than any measure of incoming freshman tech savviness (which brings us back on topic...).
I even chose CS over History and other things I was roughly equally interested in because it better mapped to a better career. (Also because my grades were stronger in CS and advanced math, but that had more to do with my odds of acceptance.) I had lots of classmates who were horrible at math but had chosen the program for the money it represented. Most of them failed out and migrated to the business program (which was less academically rigid at that school at that time; these days, they'd fail there too).
When I went off to college, many of my most IT-savvy freshman colleagues were versed in networks and system administration because they had run the computer labs of their high schools. Some of them had been caught cracking or otherwise mucking about in ways that the school staff lacked the ability to revert and been forced to clean up after themselves, others saw messes and volunteered to help out. They got paid and had responsibilities. From this new perspective, they learned the "damage" students could deal and then had the hands-on task of cleaning it up. I wish I had had that opportunity.
In this sort of environment, especially given the ubiquity of virtual machines and virtual networks, a well-facilitated capture the flag (CTF) event should be easy enough to facilitate. Even without virtualization (or even any lab at all), any school could reach out to a local hacker group and ask them to host a CTF event. The cost of scrounging up a bunch of computers and networking equipment for a one-shot event should be decently low given the spare parts in your typical hacker group or Linux users group. Maybe the school or city could even provide a budget for the event.
"The island is 1,250 km (780 mi) long and 191 km (119 mi) across its widest points and 31 km (19 mi) across its narrowest points.[1] The largest island outside the main island is the Isla de la Juventud (Isle of Youth) in the southwest, with an area of 2,200 km2 (850 sq mi)." http://en.wikipedia.org/wiki/Geography_of_Cuba
Sorry, Slashdot killed my squared symbol and I missed it in the content preview. Wikipedia says Cuba is 110 km^2 in area.
If 802.22 can cover a 100km radius (200km diameter), width isn't an issue. The 1,250km length would need only seven full-powered 802.22 antennae to provide a "backbone" across the main island (1250/200 = 6.25). Maybe each of those can have either a satellite uplink or a wired connection. Surely, another few hundred cheaper and/or lower-powered antennae (perhaps 802.11y or 802.11af?) would be able to saturate valleys and high density areas.
If Cuba built its own onion routing network (perhaps using Tor software though not connected to the Tor network), then each satellite dish or other internet connection would automatically be able to facilitate connectivity for the rest of the network. No need to wire anything (except some of the exit nodes), this can all happen over wifi.
Don't forget that 802.11af, 802.11y, and 802.22 have ranges measured in miles (802.22 can cover 100km). Blanketing an island of 110km would still take a good number of antennae (especially given the dead zones created by dense buildings in cities), but at a governmental budget scale, it seems quite feasible.
Thanks for taking the time for this, Soulskill (et al).
I really missed the ability to set comment thresholds in the GET of an article (removed in the last major UI upgrade). I have a lot of friends that do not frequent slashdot, and when I link them an article that I want them to read the better comments of, it needs to be at a threshold they'll tolerate (typically, 5/4 for full/abbrev if there are enough comments).
I have other suggestions as well, but getting comments right is by far #1. I can fix the rest with Greasemonkey.
If we want autonomous vehicles to be maximally efficient, this has to happen. They move out of the way for a police officer or for somebody who has to change their route at the last minute and get to an exit from the opposite lane. More importantly, self-driving cars can cluster together. Take India for example; they drive 4-5 cars wide in lanes marked for 3. Highly efficient, but highly unsafe for human operators.
This doesn't have to invade our privacy or be implemented insecurely. If range is limited and details forgotten when they become irrelevant, then we're fine. If cars generate random IDs, there's no way to collate them together over time (well, without existing technology like reading plates or RFID).
Wine: an emulator of the win32 API+ABI on POSIX+X.
WinXP/Vista/7/8: an emulator of the win32 API+ABI on NT.
Neither is native in this sense.
That reminds me of this gem from the Cygwin FAQ (through Dec 2009, since removed for political correctness):
Windows 9x: n. 32 bit extensions and a graphical shell for a 16 bit patch to an 8 bit operating system originally coded for a 4 bit microprocessor, written by a 2 bit company that can't stand 1 bit of competition.
The wayland repository continues to mature and moves slowly. This
cycle again only saw a few wayland changes, most of which where fairly
unexciting:
- SHM Buffer SIBGUS protection. We added and couple of utility
functions to help compositors guard against broken or malicious
clients who could truncate the backing file for shm buffers and
thus trigger SIGBUS in the compositor (Neil Roberts).
- Subsurfaces protocol moved to wayland repo and as such promoted to
official wayland protocol (Pekka Paalanen).
- wl_proxy_set_queue() can take a NULL queue to reset back to default
queue. (Neil Roberts).
- A few bug fixes, in particular, I'd like highlight the fix for the
race between wl_proxy_create() and wl_proxy_marshal().
- A few scanner error message improvements and documentation tweaks
and polish.
I'm hoping the Maui Project (which uses Wayland) can continue to gain momentum as Wayland does and that it becomes a viable option in the next few years.
From what I can tell, GCC is still the better compiler. It is better supported (lots of things won't work on clang or llvm-gcc) as well. LLVM/Clang tends to compile a bit faster (which doesn't matter unless it's an order of magnitude) while the binaries that GCC produces tend to run more efficiently. There's a nice benchmark comparing GCC 4.7 to Clang 3.1 (in Apr 2012) which demonstrates this.
I'm sure LLVM has been well designed and perhaps can do better with JIT and similar concepts (which you'd have to compare to e.g. GNU Lightning), but GCC is still king. Stallman's complaint is that it's getting attention and therefore it may become better than GCC over time, which he argues would be bad for developers on the assumption that eventually a game-changing feature is released for LLVM that is nonfree and then everybody will be forced to pay for it, a fate that the GPL'd GCC cannot suffer.
This will help find exact matches by exif data. It will not find near-matches unless they have the same exif data. If you want that, good luck. Geeqie has a find-similar command, but it's only so good (image search is hard!). Apparently there's also a findimagedupes tool available, see comments above (I wrote this before seeing that and had assumed apt-cache search had already been exhausted).
I would write a script that runs exiftool on each file you want to test. Remove the items that refer to timestamp, file name, path, etc. make a md5.
Something like this exif_hash.sh (sorry, slashdot eats whitespace so this is not indented):
#!/bin/sh
for image in "$@"; do
echo "`exiftool |grep -ve 20..:..: -e 19..:..: -e File -e Directory |md5sum` $image"
done
If you have a really large list of images, do not run this through sort. Just pipe it into your output file and sort it later. It's possible that the sort utility can't deal with the size of the list (you can work around this by using grep '^[0-8]' output |sort >output-1 and grep -v '^[0-8]' output |sort >output-2, then cat output-1 output-2 > output.sorted or thereabouts; you may need more than two passes).
There are other things you can do to display these, e.g. awk '{print $1}' output |uniq -c |sort -n to rank them by hash.
On Debian, exiftool is part of the libimage-exiftool-perl package. If you know perl, you can write this with far more precision (I figured this would be an easier explanation for non-coders).
Each employee takes, at most, one bus each way every day. Google sends buses back and forth all day long, even into the night and weekends. Yahoo and others have separate buses rather than pooling resources, so there are plenty of wasted seats compared to a unified bus system.
As to congestion: Buses can't maneuver as well as cars, and cars tend not to roost next to bus stops waiting for their turn (even if they did, they'd fit, rather than blocking traffic nearby), so yes, they cause congestion. Read the Kos article for further detail.
So they're being too eco-friendly with the bus rides? Or everyone's jealous about the benefits? Or public transportation isn't crowded enough? I don't get it but I have the sneaking suspicion that these people are morons.
I think you've missed the point. Dozens of companies in the peninsula have their own dedicated bus lines. The bus-to-person ratio is quite high, and this is not as eco-friendly as you might think. It also causes congestion in the city, and confusion at the shared bus stops (which are owned by the city of SF), both of passengers and of citizens looking for a bus they can actually ride.
The city taxing the bus services allows maintenance to be applied to the extra load of the stops as well as planning for the increased traffic these systems create. I think it is quite reasonable.
So you agree with the OP on length and quality and "bloginess", but you suggest that Bennet disregards those comments?
No. While there is room for improvement, the article is good and Bennett is not "shitting up/." I was suggesting that Bennett disregard the highly negative tone of that comment. I did not say that the article was perfect or that the criticisms of this thread were entirely without merit.
While I agree that "length does not equal insight," I think that there is insight in the article and that its length is fine. Sure, it could benefit from more concision, but most articles fall in that category. The prose is a bit "bloggy" but not unacceptably. Again, there is room for improvement but that's not enough to make this a bad read.
ML is the way to go, the trouble is that it's really really hard to do. I like the idea of having users categorize items so that you can use the hand-classified data to train on and then scale up with machine learning, but that's only a part of the puzzle. It's more difficult to properly curate what is and is not headline-worthy without catering to a basic popularity contest. Good luck with that, and may you continue to be optimistic!
I've always hoped that Slashdot itself could use ML to extend comment moderation: allow five points for human moderation, two for meta-moderation (given enough reinforcement), and three for a third system based on ML (trained by meta-moderation-confirmed moderation). Average the direct moderation (-1 to 5) and the indirect moderation (meta+ML, -1 to 5), adjust by +1/-1/-2 for AC/karma/user-conf, and round up or down based on achievements. Alternatively, make it a ten point system and add them rather than averaging them (fold achievements into karma). I'd start with the ML system as a moderator within the current system, then once it's proven, migrate to the averaging system, then migrate to the ten point scale.
I don't know what kind of system of black mail has given you the power to turn/. into your personal blog, but please stop using it like one. Length does not equal insight, your posts are not more or less important than those of other users, stop shitting up/.
Bennett, please disregard that. People do like summaries and quick reads, which is what the quoted first paragraph you provided delivers. Slashdot's audience is a little too accustomed to having to click on links to see the real article ("mindless link propagation"). Coupling that with the fact that nobody actually RTFA, you get comments like what we see above.
Frankly, I'm happy to see original content on Slashdot (well, beyond book reviews and Ask Slashdot). Thank you for contributing a real story directly to this site rather than posting it elsewhere and linking it in a Slashdot article.
(That said, I do agree with krauch aum that "length does not equal insight," I just happen to have differed in opinion about whether this article has insight. I'd also agree that this reads a little more like a blog than I'd personally like; I'm happier with items that are more like news articles than op-eds. I'd still rate this as a good write-up overall.)
The problem is that they're trying to make the manufacture process cheaper in order to eek out a slightly higher profit. Plus, multitouch is increasing the amount of space necessary to devote to the touchkpad.
I really like the keyboard on my Thinkpad X201s, whose only isue is the Escape key being above the F1 rather than to the left of it, but at least it's at the upper-leftmost corner of the keyboard, so it was easy to adapt to (also, I mapped F1 to Esc in a lot of programs). The other problem with this laptop was the small touchpad. It supported multitouch, but it's tiny, measuring about two inches wide by one inch tall.
Clearing a row of keys obviously translates to increasing the vertical size of the touchpad. I get that, it just needs to be done more intelligently. The X1's touchkpad is about four rows of keys tall and has no buttons. The X201s's touchkpad, including keys, is about 3 1/3 rows tall. There are two rows of function keys up top which are each about two thirds of a row tall. Removing one of them solves the space problem to the millimeter. Don't get rid of those keys though! I think I could survive with that top (function) key row turned into a capacitive row the way the X1 does... so long as there is still a normal Esc key on its far left (above the tilde, which is to the left of the digits).
How about:
- Backspace must be the rightmost key on the digits row.
- Esc must be the upper-leftmost key, one row above the digits' row.
- Power is fine at the upper-rightmost key.
- Also retain the standard positions for tilde (left of digits), CapsLk (under Tab).
- The top row can remain "adaptive" as long as its end keys, Esc and Power, are real.
- Give a non-multitouch middle-click option, either via emulation (right+left) or three spots at the touchpad's top.
- Grow down instead of up:
- Bring back the numeric keypad's 0-9 and dot (default NumLk off) for 4 rows of arrows/numbers. This places PgUp/PgDn, Home/End, and Ins/Del.
- New buttons (like Fn and Compose) can go under each Alt key for easy thumb access
- Fn is a hardware button, but if you add Compose, it's software (allow setting it as another bucky bit like Hyper or Option)
I've mocked this up and posted it in ascii art (mostly to scale) to pastebin at http://pastebin.com/sEECJKrh. If any laptop keyboard designers are reading this, it's free for you to use. I just want it to be used.
Hm, interesting; According to Copyright.gov, games cannot be protected under US copyright law. I had thought the same applied to patenting game mechanics, but I've lost my reference to that one.
Google is funding this (both the direct research on FreeBSD and the port to the Linux kernel) because it addresses an aspect one level above the browser. Google Chrome would then be quite tightly sandboxed. This sure beats my method of running browsers as another user (I symlink ~/Downloads to my web user's version of that area and move things out of it quickly), especially since my method wouldn't do anything against actual privilege escalation (to root).
This should also help fight web server exploits (among other malware), especially w.r.t. installing rootkits. Should be great ammunition for advocating *Nix over Windows 8.
I'm excited, but I can wait for it to hit Linux (probably; FreeBSD will release with it by default in few weeks).
I've done this in the past. My routing computer's wifi has never been able to compare with a wifi router, but if you ignore the "wan" port and plug your linux box into one of the other ports, you can use the wifi router for wifi only (essid, etc) and your own router for how traffic flows to the internet and to your wired network. The best of both worlds.
https://epic.org/ is EPIC, the Electronic Privacy Information Center, a stalwart defender of online privacy. EPIC does not appear to have any connection to this browser. This so-called "epic browser" doesn't look like much more than Iron, which was merely a ploy to make money off of ads on the download page. I'm not saying Epic Browser is that same ploy, but the browser doesn't really do anything that Chromium doesn't already do in Incognito mode (most of those 11 potential privacy leaks that epic blocks are Google features not available in Chromium or else can be disabled trivially).
This introduces a potential lag time in security updates (and updates to trackers pulled in from e.g. adblock or noscript) and rides on EPIC's good name. Shame on the developers for naming it so similarly.
Slashdot Mirror
The point is we need people like him to remind us that certain things suck and need to be replaced (cough, X11, cough) otherwise we ae stuck with old badly architected technology for decades.
It's hard to find somebody that says X11 doesn't suck. I am definitely not that person.
My point was that he says forking sucks, he gave an example where (unbeknownst to him?) forking was certainly the best option, then he went on to talk about how forked Linux distributions have made the world a better place. He seems to conclude that forking is great and that he "loves" it.
(Also, I misspelled his name. Sorry, Bryan. I guess my post had room for improvement ... meaning it sucked.)
OpenSSL doesn't listen to bug reports. They don't even accept offered patches to known bugs. It's this spirit of non-cooperation that caused the forking into LibreSSL. See the 30 day prospectus (/. coverage) from the LibreSSL project lead, which details all of the problems. Brian even says forking is ultimately a benefit, and that he "loves that they're doing it."
It seems to be that his definition of "sucks" is "has room for improvement" ... Everything has room for improvement, so apparently everything sucks.
https://s3.amazonaws.com/img.docstoc.com/thumb/orig/170154030.png
I'd paste the whole text here if it weren't an image.
But part of me wonders if I'm missing the point, if they're so intent on breaking it then might I as well just move browsers now? If I'm having to rely on addons to make a browser work then am I not just sat precariously one step away from Mozilla deciding that addon is unacceptable and cancelling it anyway?
It appears the FF devs have forgotten that their main advantage over Chrome is addons. I have so many addons, with icons to control them in the status bar (addons bar) that the new FF gave me about an inch of locationbar to see URLs. Thanks guys. I reverted this by using the dev version of Status-4-Evar. The GP's mention of Classic Theme Restorer is interesting, but I worry about its compatibility with Tab Mix Plus and other addons, as well as to your point of perhaps trying the new look & feel.
The core tenant behind the increasingly popular paleo diet is that food has been over-domesticated, favoring things like size, portability, and crop yield rather than health. Taste is often also low on the priority list (though higher than health). Wild plants like dandelion greens and ramps are significantly healthier than our domesticated cabbages for example.
The same goes for meat. Wild game meat is far healthier than meat from a factory farm. It's often tastier as well, though the farmed stuff tends to be fattier (and fat equals flavor). I'd love to try the meat of an ancestor of the cow that pre-dates its domestication. (It should also be eating and excersizing similar to the way it would in the wild rather than eating corn and living in tight quarters.)
Times have changed - when I did my computer science degree, most of the students were at the geeky end of the spectrum and were there because that's what they were really into. Compare to the present-day cross section of computer science students: most of them are there because computers are seen as a good career. The extra-curricular interest is giving way to people who just want a job.
I disagree. People like you and me merely congregated together and ignored the others. (Also, you went to school in Wales. Different world.) My above statement was about "my most IT-savvy freshman colleagues," which is to say under a dozen total (and I was friends with all of them). I'd say about 75% of my freshman peers in CS declared the major for its salaries and/or a passion for video games. I imagine today's breakdown is roughly the same, more due to the fact that most freshmen are blank slates than any measure of incoming freshman tech savviness (which brings us back on topic...).
I even chose CS over History and other things I was roughly equally interested in because it better mapped to a better career. (Also because my grades were stronger in CS and advanced math, but that had more to do with my odds of acceptance.) I had lots of classmates who were horrible at math but had chosen the program for the money it represented. Most of them failed out and migrated to the business program (which was less academically rigid at that school at that time; these days, they'd fail there too).
When I went off to college, many of my most IT-savvy freshman colleagues were versed in networks and system administration because they had run the computer labs of their high schools. Some of them had been caught cracking or otherwise mucking about in ways that the school staff lacked the ability to revert and been forced to clean up after themselves, others saw messes and volunteered to help out. They got paid and had responsibilities. From this new perspective, they learned the "damage" students could deal and then had the hands-on task of cleaning it up. I wish I had had that opportunity.
In this sort of environment, especially given the ubiquity of virtual machines and virtual networks, a well-facilitated capture the flag (CTF) event should be easy enough to facilitate. Even without virtualization (or even any lab at all), any school could reach out to a local hacker group and ask them to host a CTF event. The cost of scrounging up a bunch of computers and networking equipment for a one-shot event should be decently low given the spare parts in your typical hacker group or Linux users group. Maybe the school or city could even provide a budget for the event.
"The island is 1,250 km (780 mi) long and 191 km (119 mi) across its widest points and 31 km (19 mi) across its narrowest points.[1] The largest island outside the main island is the Isla de la Juventud (Isle of Youth) in the southwest, with an area of 2,200 km2 (850 sq mi)." http://en.wikipedia.org/wiki/Geography_of_Cuba
Sorry, Slashdot killed my squared symbol and I missed it in the content preview. Wikipedia says Cuba is 110 km^2 in area.
If 802.22 can cover a 100km radius (200km diameter), width isn't an issue. The 1,250km length would need only seven full-powered 802.22 antennae to provide a "backbone" across the main island (1250/200 = 6.25). Maybe each of those can have either a satellite uplink or a wired connection. Surely, another few hundred cheaper and/or lower-powered antennae (perhaps 802.11y or 802.11af?) would be able to saturate valleys and high density areas.
If Cuba built its own onion routing network (perhaps using Tor software though not connected to the Tor network), then each satellite dish or other internet connection would automatically be able to facilitate connectivity for the rest of the network. No need to wire anything (except some of the exit nodes), this can all happen over wifi.
Don't forget that 802.11af, 802.11y, and 802.22 have ranges measured in miles (802.22 can cover 100km). Blanketing an island of 110km would still take a good number of antennae (especially given the dead zones created by dense buildings in cities), but at a governmental budget scale, it seems quite feasible.
Thanks for taking the time for this, Soulskill (et al).
I really missed the ability to set comment thresholds in the GET of an article (removed in the last major UI upgrade). I have a lot of friends that do not frequent slashdot, and when I link them an article that I want them to read the better comments of, it needs to be at a threshold they'll tolerate (typically, 5/4 for full/abbrev if there are enough comments).
I have other suggestions as well, but getting comments right is by far #1. I can fix the rest with Greasemonkey.
I see this as inevitable, really.
If we want autonomous vehicles to be maximally efficient, this has to happen. They move out of the way for a police officer or for somebody who has to change their route at the last minute and get to an exit from the opposite lane. More importantly, self-driving cars can cluster together. Take India for example; they drive 4-5 cars wide in lanes marked for 3. Highly efficient, but highly unsafe for human operators.
This doesn't have to invade our privacy or be implemented insecurely. If range is limited and details forgotten when they become irrelevant, then we're fine. If cars generate random IDs, there's no way to collate them together over time (well, without existing technology like reading plates or RFID).
Wine: an emulator of the win32 API+ABI on POSIX+X. WinXP/Vista/7/8: an emulator of the win32 API+ABI on NT.
Neither is native in this sense.
That reminds me of this gem from the Cygwin FAQ (through Dec 2009, since removed for political correctness):
Windows 9x: n. 32 bit extensions and a graphical shell for a 16 bit patch to an 8 bit operating system originally coded for a 4 bit microprocessor, written by a 2 bit company that can't stand 1 bit of competition.
The wayland repository continues to mature and moves slowly. This cycle again only saw a few wayland changes, most of which where fairly unexciting:
- SHM Buffer SIBGUS protection. We added and couple of utility functions to help compositors guard against broken or malicious clients who could truncate the backing file for shm buffers and thus trigger SIGBUS in the compositor (Neil Roberts).
- Subsurfaces protocol moved to wayland repo and as such promoted to official wayland protocol (Pekka Paalanen).
- wl_proxy_set_queue() can take a NULL queue to reset back to default queue. (Neil Roberts).
- A few bug fixes, in particular, I'd like highlight the fix for the race between wl_proxy_create() and wl_proxy_marshal().
- A few scanner error message improvements and documentation tweaks and polish.
I'm hoping the Maui Project (which uses Wayland) can continue to gain momentum as Wayland does and that it becomes a viable option in the next few years.
From what I can tell, GCC is still the better compiler. It is better supported (lots of things won't work on clang or llvm-gcc) as well. LLVM/Clang tends to compile a bit faster (which doesn't matter unless it's an order of magnitude) while the binaries that GCC produces tend to run more efficiently. There's a nice benchmark comparing GCC 4.7 to Clang 3.1 (in Apr 2012) which demonstrates this.
I'm sure LLVM has been well designed and perhaps can do better with JIT and similar concepts (which you'd have to compare to e.g. GNU Lightning), but GCC is still king. Stallman's complaint is that it's getting attention and therefore it may become better than GCC over time, which he argues would be bad for developers on the assumption that eventually a game-changing feature is released for LLVM that is nonfree and then everybody will be forced to pay for it, a fate that the GPL'd GCC cannot suffer.
This will help find exact matches by exif data. It will not find near-matches unless they have the same exif data. If you want that, good luck. Geeqie has a find-similar command, but it's only so good (image search is hard!). Apparently there's also a findimagedupes tool available, see comments above (I wrote this before seeing that and had assumed apt-cache search had already been exhausted).
I would write a script that runs exiftool on each file you want to test. Remove the items that refer to timestamp, file name, path, etc. make a md5.
Something like this exif_hash.sh (sorry, slashdot eats whitespace so this is not indented):
#!/bin/sh
for image in "$@"; do
echo "`exiftool |grep -ve 20..:..: -e 19..:..: -e File -e Directory |md5sum` $image"
done
And then run:
find [list of paths] -typef -print0 |xargs -0 exif_hash.sh |sort > output
If you have a really large list of images, do not run this through sort. Just pipe it into your output file and sort it later. It's possible that the sort utility can't deal with the size of the list (you can work around this by using grep '^[0-8]' output |sort >output-1 and grep -v '^[0-8]' output |sort >output-2, then cat output-1 output-2 > output.sorted or thereabouts; you may need more than two passes).
There are other things you can do to display these, e.g. awk '{print $1}' output |uniq -c |sort -n to rank them by hash.
On Debian, exiftool is part of the libimage-exiftool-perl package. If you know perl, you can write this with far more precision (I figured this would be an easier explanation for non-coders).
Each employee takes, at most, one bus each way every day. Google sends buses back and forth all day long, even into the night and weekends. Yahoo and others have separate buses rather than pooling resources, so there are plenty of wasted seats compared to a unified bus system.
As to congestion: Buses can't maneuver as well as cars, and cars tend not to roost next to bus stops waiting for their turn (even if they did, they'd fit, rather than blocking traffic nearby), so yes, they cause congestion. Read the Kos article for further detail.
So they're being too eco-friendly with the bus rides? Or everyone's jealous about the benefits? Or public transportation isn't crowded enough? I don't get it but I have the sneaking suspicion that these people are morons.
I think you've missed the point. Dozens of companies in the peninsula have their own dedicated bus lines. The bus-to-person ratio is quite high, and this is not as eco-friendly as you might think. It also causes congestion in the city, and confusion at the shared bus stops (which are owned by the city of SF), both of passengers and of citizens looking for a bus they can actually ride.
The city taxing the bus services allows maintenance to be applied to the extra load of the stops as well as planning for the increased traffic these systems create. I think it is quite reasonable.
Daily Kos had a good explanation of the problem back in April.
So you agree with the OP on length and quality and "bloginess", but you suggest that Bennet disregards those comments?
No. While there is room for improvement, the article is good and Bennett is not "shitting up /." I was suggesting that Bennett disregard the highly negative tone of that comment. I did not say that the article was perfect or that the criticisms of this thread were entirely without merit.
While I agree that "length does not equal insight," I think that there is insight in the article and that its length is fine. Sure, it could benefit from more concision, but most articles fall in that category. The prose is a bit "bloggy" but not unacceptably. Again, there is room for improvement but that's not enough to make this a bad read.
ML is the way to go, the trouble is that it's really really hard to do. I like the idea of having users categorize items so that you can use the hand-classified data to train on and then scale up with machine learning, but that's only a part of the puzzle. It's more difficult to properly curate what is and is not headline-worthy without catering to a basic popularity contest. Good luck with that, and may you continue to be optimistic!
I've always hoped that Slashdot itself could use ML to extend comment moderation: allow five points for human moderation, two for meta-moderation (given enough reinforcement), and three for a third system based on ML (trained by meta-moderation-confirmed moderation). Average the direct moderation (-1 to 5) and the indirect moderation (meta+ML, -1 to 5), adjust by +1/-1/-2 for AC/karma/user-conf, and round up or down based on achievements. Alternatively, make it a ten point system and add them rather than averaging them (fold achievements into karma). I'd start with the ML system as a moderator within the current system, then once it's proven, migrate to the averaging system, then migrate to the ten point scale.
I don't know what kind of system of black mail has given you the power to turn /. into your personal blog, but please stop using it like one. Length does not equal insight, your posts are not more or less important than those of other users, stop shitting up /.
Bennett, please disregard that. People do like summaries and quick reads, which is what the quoted first paragraph you provided delivers. Slashdot's audience is a little too accustomed to having to click on links to see the real article ("mindless link propagation"). Coupling that with the fact that nobody actually RTFA, you get comments like what we see above.
Frankly, I'm happy to see original content on Slashdot (well, beyond book reviews and Ask Slashdot). Thank you for contributing a real story directly to this site rather than posting it elsewhere and linking it in a Slashdot article.
(That said, I do agree with krauch aum that "length does not equal insight," I just happen to have differed in opinion about whether this article has insight. I'd also agree that this reads a little more like a blog than I'd personally like; I'm happier with items that are more like news articles than op-eds. I'd still rate this as a good write-up overall.)
The problem is that they're trying to make the manufacture process cheaper in order to eek out a slightly higher profit. Plus, multitouch is increasing the amount of space necessary to devote to the touchkpad.
I really like the keyboard on my Thinkpad X201s, whose only isue is the Escape key being above the F1 rather than to the left of it, but at least it's at the upper-leftmost corner of the keyboard, so it was easy to adapt to (also, I mapped F1 to Esc in a lot of programs). The other problem with this laptop was the small touchpad. It supported multitouch, but it's tiny, measuring about two inches wide by one inch tall.
Clearing a row of keys obviously translates to increasing the vertical size of the touchpad. I get that, it just needs to be done more intelligently. The X1's touchkpad is about four rows of keys tall and has no buttons. The X201s's touchkpad, including keys, is about 3 1/3 rows tall. There are two rows of function keys up top which are each about two thirds of a row tall. Removing one of them solves the space problem to the millimeter. Don't get rid of those keys though! I think I could survive with that top (function) key row turned into a capacitive row the way the X1 does ... so long as there is still a normal Esc key on its far left (above the tilde, which is to the left of the digits).
How about:
I've mocked this up and posted it in ascii art (mostly to scale) to pastebin at http://pastebin.com/sEECJKrh. If any laptop keyboard designers are reading this, it's free for you to use. I just want it to be used.
Actually, board games, food recipes, and fashion designs are all patentable. Board Games: http://www.ipwatchdog.com/2011/12/22/patenting-board-games-101/id=21356/
Hm, interesting; According to Copyright.gov, games cannot be protected under US copyright law. I had thought the same applied to patenting game mechanics, but I've lost my reference to that one.
Google is funding this (both the direct research on FreeBSD and the port to the Linux kernel) because it addresses an aspect one level above the browser. Google Chrome would then be quite tightly sandboxed. This sure beats my method of running browsers as another user (I symlink ~/Downloads to my web user's version of that area and move things out of it quickly), especially since my method wouldn't do anything against actual privilege escalation (to root).
This should also help fight web server exploits (among other malware), especially w.r.t. installing rootkits. Should be great ammunition for advocating *Nix over Windows 8.
I'm excited, but I can wait for it to hit Linux (probably; FreeBSD will release with it by default in few weeks).
I've done this in the past. My routing computer's wifi has never been able to compare with a wifi router, but if you ignore the "wan" port and plug your linux box into one of the other ports, you can use the wifi router for wifi only (essid, etc) and your own router for how traffic flows to the internet and to your wired network. The best of both worlds.
https://epic.org/ is EPIC, the Electronic Privacy Information Center, a stalwart defender of online privacy. EPIC does not appear to have any connection to this browser. This so-called "epic browser" doesn't look like much more than Iron, which was merely a ploy to make money off of ads on the download page. I'm not saying Epic Browser is that same ploy, but the browser doesn't really do anything that Chromium doesn't already do in Incognito mode (most of those 11 potential privacy leaks that epic blocks are Google features not available in Chromium or else can be disabled trivially).
This introduces a potential lag time in security updates (and updates to trackers pulled in from e.g. adblock or noscript) and rides on EPIC's good name. Shame on the developers for naming it so similarly.