Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Re:Shitposting (Score 1) 3

Nope, but...

Another investigation from James O'Keefe Uncovers Plot to Chain the Trains & Shut Down DC During Inauguration

part 1
https://www.youtube.com/watch?...

part 2
https://www.youtube.com/watch?...

Okay, so there was an uptick in hate crimes, but they'll need to turn their Patented Weathercock Hate Crime Detector to point 180 degrees from its current orientation. Cuz right now they're doin' it wrong.

(still laughing at the AC's response :)

Comment Re:Smart quotes break technical content (Score 2) 207

I just had occasion to look into the reason why curly quotes often go the wrong way (and grow spurious spaces around themselves), and it's because of a quirk (or bug) in how RTF (and exported-from-RTF) handles nesting for formatting codes.

Smart quotes depend on finding ON and OFF codes with a single block of formatting, but RTF likes to put paragraph breaks INSIDE the nearest adjacent paired formatting or on/off code. Which means the parser can't find the OFF code so it uses another ON code, and the user gets curly quotes pointing the wrong direction.

Same thing with smart single quotes.

Basically, it's bad tag nesting.

Once I got to really examining all the various cases, it was clear it was all one problem that can be triggered by any change in formatting including line breaks, but may look different depending on what else is adjacent, especially when there also a code with multiple manifestations, like that for the M-dash (which has two possible codes, and behaves differently depending on whether there are trailing spaces).

Every RTF editor and export-to-RTF I looked at had the same problem. So it's probably a failing of the RTF standard (such as that is... so many to choose from!) that sorta neglected to specify how code nesting must be handled.

Comment Impalpable evidence. (Score 1) 404

I read the report. It's extremely vague; mostly quotes a Microsoft document on generally securing your shit. It doesn't actually put forth ANY hard evidence of ANY hack, Russian or otherwise -- closest it comes is citing a snippet of source code of unverified origin.

I hate to cite McAfee since he's such a nutjob, but I think he's completely correct here:

http://www.express.co.uk/news/...
====
"Any hacker capable of breaking into something is extraordinarily capable of hiding their tracks. If I were the Chinese and I wanted to make it look like the Russians did it I would use Russian language within the code. I would use Russian techniques of breaking into organisations so there is simply no way to assign a source for any attack -- this is a fallacy." He argued the report was part of a ploy to "manipulate our opinions".

Comment Re:Tiger repellent (Score 1) 277

That cone thing is nonsense, but adding solar panels to windmill blades? That might be a good idea -- for one thing it gets them up out of the surface blowing dirt that etches hell out of the panels. For another the rest of the unit is already budgeted. So even if the solar isn't angled efficiently, it could be a bonus.

Comment Re:Why they are slow? (Score 1) 766

I use HOSTS (that one, in fact), NoScript, and PrefBar (to have control over flash, acrobat, image loading, etc.) and that pretty much strips what I get in my browser down to the minimum to make a given page work. On average my browser loads maybe 10% of what the server would like to send me. (No adblocker required, either.)

But even thus, modern browsers are still slow, nothing to do with the server or devs' finest efforts at optimization; as best I (not a coder but an experienced observer) can tell the problem really is in the browser itself: If you run any of the Mozilla family on a slower machine, you can actually see it doing linear singletasking (most visibly when asked to save several files in a row). Old Netscape didn't have this problem; it arrived with the opensource version. One suspects a plague of bad coding Zen [a la Abrash's Black Book].

And why does it take 100mb of RAM just to display a 10k text-only page?? and up to 1.9GB to run a couple videos? the same HTML renders in any other program at a cost of a couple MB, and flash itself uses very little. My observation is that the problem is mostly in plugin-container.exe, but it's not the whole issue. Disabling disk cache entirely helps too (its current convoluted structure seems to cause a major memory leak -- seriously, 1500+ directories holding one file apiece??), but not enough.

The latest bizarre slowdown (this is in SeaMonkey 2.39, can't run a newer one BECAUSE IT'S TOO DAMN SLOW) is running CPU at 100% for 30+ seconds when asked to merely copy text from a webpage. Thank ghu for the "Copy as Plain Text" add-on, which works around the problem.

Performance gets significantly worse with every update, which is why I've fallen into the practice of using the oldest browser that will still halfway work.

Browser devs need to be restricted to working on old hardware, so they can see what they're doing to the rest of us who can't justify new hardware just so the browser doesn't take forever to do its job.

Slashdot Top Deals

We all like praise, but a hike in our pay is the best kind of ways.

Working...