Slashdot is powered by your submissions, so send in your scoop


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

Comment Only if it seems slow for the intended use-case. (Score 1) 218

Donald Knuth's statements on premature optimisation come to mind here. If the code seems to work in a reasonable timeframe, I don't even bother. If it seems a bit slow (or to use a bit too much CPU), I'll profile it and consider it if it's a narrow, "hot" code section. If profiling shows execution all over the place, I'll probably just rewrite the manual to require higher specs instead.

Comment My biased two cents. (Score 1) 338

Biased because I personally know the author, but... give Reddened Wasteland a shot. It's by an indie author I personally know, Kyle Perkins. Cool dude, and it's quite an interesting sci-fi dystopic, the first part of a trilogy, and it's a bit like Red Mars, just a bit more character-driven. Look for it online or in your local library with ISBN 978-1523716753.

Comment Re:Lol no (Score 1) 158

Quick Charge. People want their phones to top up from dead in less than an hour. Forcing that much current into a lithium-ion battery isn't a good idea.

The solution? Make them take longer to charge. Puts less stress on a battery that's made out of volatile materials. Stop trying to appease impatient braindead idiots that want a design feature that's inherently incompatible with safety w.r.t. current manufacturing materials for batteries.

It's been known for several decades now that charging a lithium-ion cell too quickly = ALLAHU AKBAR!, yet people still want it, why? Because they choose to smoke weed and get pregnant instead of paying attention in high school chemistry class, and grow up to be fucking idiots.

Comment One website I know to be whitelisted... (Score 1) 337 A long time ago I had a 30MB data plan and this was one of a few websites that continued to work after running out of data and getting paywalled, although most of the graphical assets were stored on a different domain and thus didn't load post-data bucket depletion. also used to work, but that was plugged in 2014.

TCP port 53 used to also be wide open, but from what I gathered on various forums, that was patched during the last major VoLTE outage. Two other users commented elsewhere on this story that this port (and it's UDP counterpart) as apparently still open on Verizon Wireless. However, I'm unable to confirm this paragraph.

Comment Re:The real reason it works: (Score 1) 337

Well, from a customer-facing perspective, speed tests on unactivated SIMs would be useful for determining what one can realistically expect for network speeds at a given location/time. A very minute "try before you buy".

Due to the way GSM/UMTS/LTE networks perform, there's no performance gains to be had in unconditional whitelisting of speedtests, since the air interface (wireless last mile) will ALWAYS be the primary bottleneck.

Comment Re:Or he could just use one of the free cell servi (Score 1) 337

That's because they initially launched on the Sprint network. Guess what, if you've ever used Verizon Wireless, the calls sound just as bad because they're using the exact same codecs and network protocols (3GPP2 CDMA2000 1xRTT). FreedomPop has partnered with a GSM operator as of late (I think T-Mobile), which provides a far superior quality of service.

Comment Re:Unauthorized access (Score 4, Interesting) 337

That whitelisting for speedtests also applies to unactivated SIMs and prepaid SIMs without active service (e.g. due to nonpayment or zero balance.)

I used to keep a spare phone lying about with an unactivated SIM while I had a prepaid SIM, and discovered the speedtest whitelisting was unconditional. I never thought to dig any deeper into it, although I suspected this type of thing was possible all along.

Glad to have my suspicions confirmed without having to risk my ass.

Comment Re:My opinion... (Score 1) 405

You should probably look into what you agreed to send Microsoft in the EULA

Surprise! I did. And maybe I'm not as paranoid as you, but seriously, this is fairly standard stuff for any software that has built-in telemetry. Do you know why telemetry is important? Because it shouldn't take four years to fix the next Windows Vista. If Microsoft fucks up a UI decision, they need to know quickly and have the changes rolled back or fixed in the next update. Not four years later.

Even if you want to push the "but Stallman said it's 1984!" argument...

  1. Stallman needs to lay off the drugs
  2. Your PC is literally the last thing the government's going after if they're going after you. Far easier to go after your communications than the device used to make said communications, with exceptions being in E2E encrypted comms.
  3. If Microsoft was really snooping all the interactions on your PC, backbone operators would notice a few gigs of traffic coming from every Windows 10 PC that wasn't there before. You think capturing just the keystrokes is enough useful data? No, you also include metadata such as typing speed, where it was typed, screenshots, etc., and that turns a few kilobytes of typed text info a few megabytes of metadata to make sense of it all. Somebody would notice something if it really was as bad as people like yourself try to argue it is.
  4. And most importantly Hanlon's razor applies to all human interaction. Never attribute to malice that which is easily explained by stupidity. And, let me tell you, people (governments included) are pretty damned stupid.

Now, run along now, and make sure to refit your tinfoil hat because Wi-Fi's giving you brain cancer that can send your throughts to the NSA or whatever paranoid conspiracy crap you're buying into.

Slashdot Top Deals

What is worth doing is worth the trouble of asking somebody to do.