Follow Slashdot stories on Twitter


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re: Considering how few boys graduate at ALL (Score 1) 355

The question is why?

If the answer is that the small subset of men who enter the teaching field teach better then cool (maybe thats the trend that these institutions have seen). However if it is just to make up diversity numbers and they are worse teachers then that is a very bad practice. I don't have kids but I would want the best teacher for them regardless of gender.

Comment: Re:So what? (Score 1) 242

by HappyPsycho (#48251911) Attached to: OEM Windows 7 License Sales End This Friday

If you change the OS and the manufacturer finds out it is possible that they will not honor the warranty seems like a good reason to not touch the pre-installed OS. Especially when you factor in the system recovery will put said OS right back (assuming you didn't kill that partition during the install of the alternate OS).

Then you open the can of worms that is drivers, quite a few Windows 8 machines lack drivers for win 7 and previous (then again makes a good argument for switching the users to Linux).

Comment: Re:Boy toy (Score 1) 786

by HappyPsycho (#48205517) Attached to: NPR: '80s Ads Are Responsible For the Lack of Women Coders

I agree with your point (assuming you are talking about a decade ago, I expect the computer access issue has changed slightly) but that doesn't counter the GP's point of the tech industry isn't the one doing that discouraging.

In the 90s (when I was a teenager) girls looked down on geeks (then again so did most of society), allot of the time computer science had the same set of questions to answer as theoretical math "What good is it? What can I do with it?" In the 90s tech was only making large effects in very limited areas. Compare that even to Civil, Electrical and Mechanical engineering; you can see the bridges, televisions and cars respectively. Only recently (say the past 5-10 years) did the effects become easily visible to the general population.

Comment: Re:Seriously? (Score 2) 533

by HappyPsycho (#47863621) Attached to: AT&T Says 10Mbps Is Too Fast For "Broadband," 4Mbps Is Enough

Or maybe if your running at 240p, 4Mbit isn't enough for 1080p (which is around 5Mbit) so even with their definition you can't enjoy youtube to its fullest either.

Dial up is (barely) enough to run a single VoIP session (assuming you are not using G711 at 80Kbps).

Of course this is all assuming a single user per connection at a time.

Comment: Re:"Best" users (Score 1) 129

by HappyPsycho (#47841319) Attached to: The Frustrations of Supporting Users In Remote Offices

For your company, remote users are the most expensive to support. It often takes several minutes to try to make the user understand what you want them to do, and to do it PROPERLY, where locally, you could just go to a user's desk and fix the problem in seconds.

Therein lies a pretty big problem IMHO, even if it takes you a couple minutes the remote user can now deal with the issue themselves (assuming its something that doesn't require co-ordination with us). Also they now have documentation (in the form of e-mail) in the event they forget. If you just walk over to the user's desk they are not going to bother remembering how to fix it themselves they will remember where your desk is the next time.

I acknowledge from a debugging perspective it can be harder but I work for an ISP, most cases are remote and we've put in lots of instrumentation to help in those circumstances. As for some of my local users, they know where my desk is by heart and almost never provide any sort of useful information in their reports.

Comment: Re:Welcome to Australia, Ferengi. (Score 1) 139

by HappyPsycho (#47797173) Attached to: Australian Consumer Watchdog Takes Valve To Court

I have a few issues with such a sentiment:

From the article Valve's policy is "was not under any obligation to repair, replace or provide a refund for a game where the consumer had not contacted and attempted to resolve the problem with the computer game developer". Whereas the assertion by ACCC states "It is a breach of the Australian Consumer Law for businesses to state that they do not give refunds under any circumstances, including for gifts and during sale".

I don't see a real issue here, Valve are essentially saying you at least need to make contact with the developer of the game first and attempt to get your issue rectified before they will step in (the "had not contacted and attempted to resolve the problem with the computer game developer" part), maybe I'm speaking a different english than Australia (or words mean different things over there) but this seems quite reasonable. If I buy something from a third party via either amazon / ebay and have an issue with it I'm pretty much required to at least speak to the supplier (through their messaging system so there is a record) before amazon / ebay get involved.

Also most of the "broken" games on steam have a common tag, "Early Access". I'm guessing if this lawsuit goes through the entire section will get locked off to Australia. Which brings up a interesting question, how well does Australian law and crowd funding get along?

Comment: Re:Gradients (Score 4, Interesting) 32

by HappyPsycho (#47574991) Attached to: NASA's JPL Develops Multi-Metal 3D Printing Process

While I doubt it is unique to 3d printing (I could be wrong though) but the simple pour into mould methods won't work without taking into account the relative densities of the metals involved (depending on how long they take to cool they may separate out anyway).

The real benefit I can see here would be from the ability to control how fast you move from one material to another which seems to be one of the major benefits (having the gentle transition of the alloy removes the transition point and the matching weak point).

What may be unique is the control that 3d printing offers, I'm sure someone can create without using 3d printing but I'm also sure its not a quick / easy process.

Comment: Re:FCC Reminds American Public... (Score 1) 38

by HappyPsycho (#47523503) Attached to: FCC Reminds ISPs That They Can Be Fined For Lacking Transparency

Given the pushback from the supreme court and lack of fuck given by congress this is actually one of the few ways the FCC can actually enforce net neutrality (or in this case force the ISPs to say up front they are not neutral, which assuming an educated public should result in lost business to the non-net neutral ISPs).

Comment: Re:I disagree (Score 3, Informative) 390

by HappyPsycho (#47483307) Attached to: Verizon's Accidental Mea Culpa

Sorry that scenario doesn't fly:

Netflix has peerings with:
        AS2828 XO-AS15 - XO Communications,US (Tier 1)
        AS55095 AS-NFLXCORP - Netflix Inc,US
        AS3257 TINET-BACKBONE Tinet SpA,DE (Tier 1)
        AS4436 AS-GTT-4436 - nLayer Communications, Inc.,US
        AS3356 LEVEL3 - Level 3 Communications, Inc.,US (Tier 1)
        AS26592 ALOG SOLUCOES DE TECNOLOGIA EM INFORMATICA S.A.,BR (Tier 2 - Has large footprint in latin america).
        AS1299 TELIANET TeliaSonera International Carrier,SE (Tier 2 - Apparently the largest fiber providers in Europe).
        AS174 COGENT-174 - Cogent Communications,US (Tier 1)

So no this isn't a case of exclusive peering, Level 3 being such a large provider just happens to be the best connection between Verizon and Netflix.

Secondly, that whole thing of 'Level3 to Verizon: "Ok, that will be $X"' has no bearing on a peering agreement, the statement would have been more like "The link between us is congested, want to upgrade the link?" each side upgrades their switch (if neccesary) and they connect the cable / fiber (given that they are in the same location we are talking about a multimode fiber patch at the high end).

Comment: Re:This makes sense. (Score 1) 280

And this is why the algorithm method won't work: people can't keep their mouths shut.

I can think of a simple counter to this statement, RSA (or any other major encryption scheme) are published algorithms whose security is decided by the strength of the keys used.

Given the GP's algorithm I see numerous points where I can choose my own keys / base passwords / padding words. These are defenses to stuff like rainbow tables and word lists and should deny an attacker any benefit over raw brute-forcing of the password (from say a hash).

For those sites which require frequent password changes and don't allow repeating passwords something as simple as tacking on a series of digits in order like "012" then "345", "678", "901", "234", etc. will make brute-forcing a password significantly more difficult (if you notice the sequence doesn't repeat immediately when you loop back around, there are 10 combinations so it has increased the difficulty by an order of magnitude). You can easily change this to other keys on the keyboard or leters of the alphabet etc to increase security without altering the algorithm itself (just like RSA the size of the keyspace dictates the security).

Adding features does not necessarily increase functionality -- it just makes the manuals thicker.