Re:Wondering
on
Dorm Storm?
·
· Score: 4, Interesting
Not really, and for several really good reasons.
First off, he says that out of the 3500 students invading the campus, 1500 of them will be screaming Mommy when they head in and try and get connected. This is about right for A) The incoming Freshmen, and B) The terminally stupid upperclassmen.
Also, the number of people bringing computers to school with them and thinking that the archaic 8088 XT that they just dug out of the basement - usually because their parents can't or won't let them take the high-end 486 that the family uses - might be a bit surprising. (This is of course an exaggeration, I hope. None of my friends who've been there and done that ever mentioned anything quite so drastic.)
Also, there's the fact that, while the NETWORK might be able to take the abuse, it's not guaranteed that the Network ADMINS can handle the stupidity. Super-cheap-laptop + Win2K + Novell + Wireless = Twitching Admin. I really can begin to imagine the hell of it all compressed into three days or so, because - even if it wasn't tech related - I've worked the bookstore during hell week at a fairly large University before. You can't begin to imagine the disruption of life that occurs to the people who work on campuses at the end of the summer unless you've been one.
And yes, I imagine he IS in fact crying over his lost phat pipe. =)
For starters, the PEERING arrangement is gone. There's a difference between peering and transit. Peering is done when it would be mutually beneficial for both parties to share traffic, and some arrangement regarding the cost of the line - in the case of a private peer such as this - is worked out between the two. Transit is where the ISP pays out the nose for a link to the Network provider's network, allowing the ISP to use the Network provider as an upstream to access most of the world.
Now, I work for a fairly large ISP, with a nationwide privately-owned fiber backbone. We get 16ms ping times from Maine to the Carolinas, and 70-80ms pingtimes from New York to San Francisco. And we've got TONS of peering arrangements. But there's three networks that we still aren't big enough to get transit with: Level3, UUNet, and Cable & Wireless.
Big network providers like C&W don't peer with the small fries, and if you have a peering arrangement with one of them, you're right near the top - Tier One or Two Network provider. If C&W is dropping peering for PSINet, that means that, in their opinion, PSINet isn't classed as a Top-Tier network provider anymore. And that is the sound of the bell tolling doom for any network provider out there. That's why this is news.
I think that there's a few people here, from a quick glance through of the posts, that are forgetting something fundamental here.
This is JAPAN. The culture CANNOT GET MUCH DIFFERENT than the United States.
Corporate Culture in Japan is something totally different than going to the bar after work with a few coworkers, and is much more akin to citizenship in a very patriotic country. Health Club, medical, dental, living spaces, restaurants, movie theaters, all of this can and in some cases usually is provided by the company for their employees. Read up on the RPG Shadowrun to get an example of Japanese Corporate Culture taken to an extreme, then realize it isn't too extreme.
Furthermore, the corporate world of Japan is very much their replacement for the warriors of old. Employment with a corporation borders on feudal vassalship, and this is not only accepted, but considered normal by many. Sitting over here stateside, where an employer wouldn't dare try a stunt like that, and saying that it should be illegal in a society that doesn't match ours is a little bit ridiculous.
Your Mileage May Vary, but that's the reason *I* see for these particular actions, and I don't see what the big deal is.
To the best of my knowledge, there have always been caucasians in the Orient, as evidenced by Japan's own "Native Japanese" population, a tribe of caucasians that were living in Japan when the first modern Japanese landed on the Islands.
Also, if these have been turning up since the 1800s, is there a major archaeological relevance to that area? What was there at the time, anyone know?
For one, it doesn't matter how many CDs you purchase a year it doesn't justify pirating ANY amount of other music. Sure, CDs are overpriced and that's not good but you are still depriving the artist of their well deserved income. Buying 30 CDs a year doesn't mean anything to an artist whos CD you -didn't- buy.
The point that he was making here was fairly simple. It was something that was pointed out on CNN a little over a year ago. The day that the RIAA announced their lawsuit against Napster, the blurb that followed it was "on a related note, CD sales are up an estimated x% this quarter" - paraphrased, and I think x was around 15 or so. There was a demonstrated increase in CD sales once Napster became popular. Whether or not it's due to Napster is another story, but I think it is.
All governing bodies have some amount of evil, and it's easy to overlook the good when all you care about is what they are depriving you of (free music).
Statement is true, concept is... flawed in my opinion. If artists were making more than pennies on the dollar for their music, I might be able to accept this as a valid argument. But this is the Recording Industry Association of America, not the Recording Artist's Association of America. RIAA exists to protect the recording studios primarily, and as a side effect, it sort of protects the companies' artists - from everyone except the companies themselves, that is.
What makes you think that Micorosoft should give away upgrades to their software, simply because you personally gauge the price to be too high? I know it's a fair whack, but to think that you are getting all that product (consider the developer's time that went in to making this stuff) and you just think you are welcome to free upgrades?
Yes.
Reasons why: I go out and buy Adaptec EZ CD Creator 4.0 so that I can burn backups of all my text files - and I have a ton of them. I go home, find out that 4.0 doesn't support my neato-nifty-keen 16x10x40 burner that I got white-boxed from some 'net retailer. I hit the EZ CD Creator Website, and I see that there's an update to version 4.02, install that, it works with my burner. I'm happy again.
Now, you're most likely to say, "Yeah, but that's a point-release, not the same." My response is, tell me one thing about Windows 98 that couldn't have been done with a point-release type of update. I really don't see it. Let's say I upgrade my old Pentium 233 to a K6-2 500, and I'm running Windows 95. Suddenly, my computer won't finish starting Windows, and it's a SOFTWARE issue. I have two choices, fall back to my Pentium 233 and hope that I can find the executable - only usable in the Windows interface, by the way, which means you HAVE to fallback or underclock - that is buried somewhere in Microsoft's website, or go out and drop a hundred bucks or more for Win 98. This is a problem I've run into recently - also affects Athlons in excess of 1GHz under Win95 - but it's not included in Windows Update, and it's so well hidden that it took me two hours to find it. This is a software bug on Microsoft's part. Why isn't it included in Windows Update?
Microsoft actually bucks industry standard practice when it comes to updates. It should cost significantly less to upgrade if there's no way to add compatibility for new hardware to the old OS. I do agree, however, that the comment about the WPA, while not two-faced per se, is a contradiction in concepts.
Everyone's entitled to an opinion I guess, but this is clearly just a college kid that's pissed he doesn't get enough pocket money.
This I think is a bit out of line, since he's really not griping about not having enough money. He's griping about software prices.
Example: I'm a network engineer. I make a fairly good living, drive a sporty type of car made in 1997, have plenty of money left over to eat with once I pay all the bills, etc. Overall, I'm at a position where I'm living comfortably. Why is it that if I wanted to go out and buy, say, Microsoft Word or Excel, or heavens forbid Visual Studio, that I would have to save up for the next six months to buy it - either that or skip the rent AND the car payment - just so I can have a decent, industry-standard word processor and spreadsheet? Some will say get StarOffice, or maybe suggest AbiWord under Linux. That doesn't work too well. You go looking for a new job, and every recruiter or headhunter you come across is going to tell you to send them a resume in Word format. You ever try writing a resume in wordpad? Not the easiest thing to do and still have it presentable.
This is also moot. Most software companies that make productivity software simply grit their teeth and ignore the individuals with pirate copies - for the most part. It's the businesses with site licenses that they need to worry about. One company is literally hundreds and thousands of dollars in revenue.
It may not be news, but then again, neither are a lot of Op-Ed pieces in a traditional newspaper. They still see press though. Why shouldn't this?
I've tried Stormix, as well as Debian Slink and Potato at various times (I do some light to moderate work on the Linux Router Project over at leaf.sourceforge.net, which is based on Debian Slink, so I need a copy of it around.) None of the above are my full-time distro - I use SuSE - but I try and install and tinker with every distro I come across.
Stormix crashed and burned on my system the three times I tried installing it, and I didn't have anything too obscure in there either. Potato and Slink are both fairly awkward to install, but you can usually get to a shell prompt easily enough.
However, I personally take a wee bit of umbrage at the "easy way to install Debian" bit in the main article, because it is most definitely NOT the only way to do it. And Slashdot is what put me onto the alternative in the first place.
In my opinion, the Holy Grail of Linux Distros, the one that does what even Mandrake (7.2, I haven't been able to get 8.0 yet) can't do for ease of install, is Progeny Debian GNU/Linux.
Note that I don't use this on a day-to-day basis, mostly because of the blood-sweat-and-tears I've put into my SuSE boxen, but this is THE single easiest-to-install distribution out there that I've come across. It pegged ALL of my hardware - even my Logitech Cordless trackball, which SuSE's quite excellent SaX and SaX2 programs could not - first go around, and was easier to install than Windows.
Yes. Easier to install than Windows. Not for us, the average geek, but for them, the average Windows user.
Now, there are a few problems/bugs with the Progeny that I played with - namely, minor bugs in the installs package selection program that causes things to go wonky if you unselect a package in Expert Install - but overall, it's smooth as hell. One of my friends at work tried it after I suggested it to him; he installed in in about half an hour, got it up and running, and grabbed Ximian Gnome, and he's happier than hell. It's based off of a snapshot of Woody, so whether the installer is Woody's or Progeny's I don't know. What I do know is, if I ever need to do a system reinstall, SuSE will only remain on my server, and Progeny will be my desktop. SuSE's been good to me, but Progeny is unbelievable.
Before you wonder, no, I don't work for Progeny. But there's quite a few Debian notables that do.
That's okay. I actually appreciate the comment more. (Not that I don't like the Karma. =)
SDSL is far more important than DSL overall, yes. Unfortunately, it's significantly less stable. As soon as they can get some more robustness into things - like, not having to worry about bad rainstorms (I've lost track of the number of DSL circuits that have gone down due to wires getting wet) then I'll be recommending SDSL to everyone.
Honestly though? I've got an ADSL line at home, and it's 1.5 down and 384 up. A lot of people don't realize that you can run a decent number of services over 384k and not have to worry about the bandwidth. It also will leave you room on the downstream for browsing, even if your server is getting smacked. So I honestly can't say that SDSL lines are necessary for services.
*Grin* I was wondering if anyone was going to ask that.
I work in NetOps myself, and I hadn't heard of a T-6 until about two months ago. Verizon uses them, at least in New England, for minor backhaul carrier circuits. From what I understand, a T-6/DS-6 is comprised of 24 DS-3 channels. Talk about your fat pipe.
To the best of my knowledge, they're rarely used anymore, and are really only leftovers from areas that haven't been fully rolled to fibre backhauls. Then again, this is Massachussetts, and we've still got a few places with pulse-dial only switches. Go figure.
DSL is not something as simple as sticking a DSLAM, and a DSLAM is actually a shared access node also. The phone wire itself must be update and in top condition, and IIRC DSLAMs can support about a T1 in total transfers at once, no matter how many people there are.
This is not correct. Yes, I'm being blunt.
The REASON this isn't correct is because of the design of a DSLAM. DSLAMs are made up of card shelves. Each card has X number of customers - I think that Covad SDSL cards allow 6, but don't hold me to that - and there are multiple cards per shelf. There's at LEAST one ATM DS-3 running off of each DSLAM into the Covad networks - Covad is what I have the experience with - and they add more pipe based on the number of users.
This CAN lead to problems, especially if 30 or so users decide to use their full bandwidth all of the time. DSL ISPs have ways of handling this, usually by warning them to stop doing it 24/7, or by shutting them down. Either way, with ATM circuits, you can have some five or six hundred people going over a single DS-3, and still only use about 20Mbit/sec worth of bandwidth on it.
As for the phone pair, it does NOT need to be up to date and in top condition; Harvard.net - one of the first DSL providers to go under, and I think exclusive to the Northeast - was buying "dry" pairs with no conditioning done to it from Verizon. DSL will run over crufty old alarm pair for fire alarms and the like; just don't expect it to stay up and running all the time. A lot of the DSL providers out there are now getting conditioned pairs to use for lines, but it's by no means necessary.
As for your brother's download speeds, that depends entirely on the ISP. If they've got an overloaded backhaul, or their transit to upstream providers is full, then peak times are going to suck. LOTS.
One person I know of is currently working on his Battlebot right now, and the basis for it is going to be a heavily modified Linux Router Project disk, using CompactFlash for the "disk", one of the Linux BIOS projects for boot, and a couple of wireless LAN cards for controls. He's keeping a radio in it for backup, but I'm betting that Battlebots won't ever be the same again.
Not to mention the fact that there's a lot more tech in the R/C and Ham Radio worlds than most purely computer geeks want to give them credit for.
Sure, why not. I've been posting more tonight than I ever have since I first started reading Slashdot. =)
For starters, there's the old saw about the shared node. With DSL, you don't need to wonder what's going to happen to your bandwidth when the rest of the neighborhood wakes up and smells the fat pipe. Another big one is, 99% of the cable providers out there throw a fairly big stink when you try and run a server over a cablemodem - once they get around to noticing it. My DSL ISP - Speakeasy - has two restrictions: No porn websites, and no IRC Servers. (And even that they can overlook if it's a single-node server and not part of a network, I think.)
In addition to that, DSL is cheaper. How? What if I don't WANT cable TV? Most cable companies won't sell you cablemodem access without a TV subscription as well. (Rare, but true. One of my friends is stuck with this dilemma.) Also, DSL is available where many times cablemodem isn't. Reverse is also true, of course, but unlike cablemodem access, you don't need to upgrade the existing infrastructure to accomodate DSL. Just stick a DSLAM in a central office and start plugging in lines.
Lastly, very few cable companies want to give you the same bandwidth upstream as you have downstream, without paying a LOT of money for it. DSL is comparatively cheap for synchronous speeds.
Hope that answers some of your questions. Incidentally, line conditions for DSL and Cable both are much better than average in Seattle, since there isn't a couple hundred years of cruft in the way.
Then you're making a mistake. Don't mean to be blunt and rude, but them's the breaks.
Your business class DSL comes with no uptime guarantees. Telcos purposely drag their feet when they have to dispatch on a DSL. Support varies from ISP to ISP badly enough. And worst of all, there is NOTHING ELSE with a lower priority than DSL when there's major storms/natural disaster outages. Priority goes like this:
Further, you will get LESS bang for your buck at equivalent speeds, because - simply put - you've got a 10% overhead built-in, since most DSL Providers use ATM-based DSL setups. 5 out of every 53 bytes of data is header information that gets stripped out by the modem or router.
If you've got a business, and you need an internet connection, get a T-1. Or even a Frame Relay. Anything else and you might as well be gambling.
The problem isn't that DSL doesn't work. It's cut-rate pricing offered by myopic providers to feed myopic customers that leads to a sudden boom followed by a bust.
This isn't necessarily true. The local telcos are legally only able to charge a minimal fee to CLECs - Competitive Local Exchange Carriers, which is technically what a DSL provider is - of around $7 for the local loop. The DSL providers then turn around and sell access to their network for X number of dollars, plus a fee per circuit of - I THINK - around $40 per line. (This is what it was about a year ago; whether or not that's changed since then, I don't know.) On top of that, you get whatever else the ISP charges. The rates aren't cut that much really; you're looking at an over 300% profit on the loop charge alone, and I know that most DSL ISPs that are worth looking at twice are charging around $60 for a 608/128 line, $90 for 1536/384 lines, and around $50 for a 192k SDSL line, working up. (Highest I've seen is $450/month for a 1.5Mbit SDSL line with some real nice side perks.)
DSL is far from cut-rate; the companies involved just happened to jump in too early to gain the early ground on it. Support costs a lot for DSL, although that's changing with the new Rate Adaptive Error Correction software out there. My line had a hard short on it at one point about two months ago, that I found out about quite by accident. My DSL line was still up through it, and - having troubleshot for umpteen million DSL circuits - I know that a hard short on a SDSL line takes it down. (In case you're wondering, the reason it isn't applied is because it would break the guarantees of speed on an SDSL line. The error correction drops your speed to compensate for loss of signal strength.)
I agree that DSL doesn't really work - currently - but I don't agree that it's because of cut-rate pricing. These companies are making up lost ground, but they're doing it slowly, because they can't afford to tinker with it to get it to work better.
Semi-useless factoid for the day. Copper-based T-1s are usually HDSL, which is the oldest of the DSL types. It's the repeaters and switching equipment - along with the cost of the loop and REALLY strict Service Level Agreements - that make all the difference in the world.
The Great Covad DSL Purge of 2001 was actually the leftovers of 2000. This was not, will not be, and all-around has VERY little negative effect on Covad.
The only purges that Covad has been doing are those where they're pulling gear out of thoroughly unprofitable COs - because, for instance, there aren't enough subscribers to afford the cage (we're talking two or three people in the CO here) - and the shutdown of service to those ISPs that are unwilling or unable to pay their bills to Covad. If anything, this made them MORE reliable as a long-term prospect. They dropped their dead weight, and their customer service at it's worst was light-years better than Northpoint's average - and still a bit better than Northpoint at it's best.
If someone got caught in a purge, then they went with a dinky little ISP that didn't pay their bills and didn't let it be known that they were in danger of losing connectivity. These providers - ALL providers that sold Covad DSL for that matter - were notified of the impending shutdowns. Covad even had a project set up to help customers transfer to new ISPs. Speakeasy was one of them - and it's still going on. (Semi-Disclaimer: I work for RCN, but Speakeasy is my DSL Provider. There's reasons, but most of it boils down to cost.)
Rule Number One of DSL: If your business relies on it's net connection, DO NOT UNDER ANY CIRCUMSTANCES USE DSL AS A PRIMARY CONNECTION! I cannot stress that enough! DSL has absolutely NO service agreements, beyond a guaranteed 80% of rated line-speed for "business" class SDSL circuits. DSL is great for the home, and good for a backup circuit, but don't think it's a T-1 for cheap.
Trust in the fact that, if ANYONE goes ANYWHERE anytime soon, it won't be Covad. They're running strong, they laid off a few people back about 6-8 months ago. After that, their support IMPROVED. They reorganized, and they're running better than ever. And their service is better, more reliable, and usually more intelligently-staffed than just about any of the local Telcos are. Do your homework on DSL, and you won't get burned.
Um, not sure where you got your information, but I'd like to provide a different view.
I do commercial Level 2 Support for a large national ISP. We had a bunch of customers through Northpoint, and the network went down hard yesterday. My personal view on why they went under is that their support was horrendous.
When your line went down - and if you have DSL, your line WILL go down someday - you call into your ISP. Your ISP verifies their side, and then called Northpoint to open a ticket. You called into Tier 1, which basically took calls, noted down information, and read back to you what was posted on their support website. This info was largely useless, because as recently as October, right around when the deal with Verizon first came up, Northpoint used a paper-only trouble ticket system. Yes. Paper Only.
If, after two or three hours, the ISP hasn't heard anything or been contacted for testing, with Covad, you call up and talk to someone who can do the testing. With Northpoint, you had no choice but to call the Tier 1 folks, who promptly told you that there was nothing that could be done because they have four hours to respond. (I once called 2 hours in, sat on hold for an hour and 45 minutes for Tier 1, and when they picked up, told me that there was nothing that could be done, then hung up on me.) No matter how far you escalated the issue within their management, no matter how long the ticket had been open with them, you could not call any sooner than four hours and expect anything to get done.
Eventually, once the deal with Verizon fell through, this began to change. Rumor has it that they started using a computerized ticketing system at some point in there, and they removed the four-hour requirement. We were provided numbers that went directly to the Tier 2 testers, and sometimes things actually got done. There were still annoyances - if a customer went down, and it was determined that their router or DSL Modem went bad, a tech was sent out, tested with a spare modem, verified it, and then took the spare modem with them when they left!!!! Issues such as this were the norm, and why Northpoint went under is that their Partner ISPs - like us, who had over a thousand DSL customers through them - were constantly given the runaround when it came to support, and they didn't realize this until they were right on the brink, with Verizon backing out and virtually no hope of actually securing investment capital.
Northpoint had bad business practices and poor support, realized it way too late to do anything about it, and paid the price. Maybe their service was cheaper and more reliable than Verizon. But don't forget that you get what you pay for, and don't make the mistake of thinking that they were a decent company.
The entire historical significance of Worcester Massachussetts is that Robert Goddard popped a rocket off a farm somewhere near it. In the past year and a half, I have yet to see anything else worth mentioning.
As for the culture, I live right in the middle of a heavy Italian neighborhood, and there's a place down the street where I can get handmade Italian sausages for the sauce that I make every so often. Some Thai/Polynesian places, some assorted other cultural backgrounds here and there, but by and large, there's nothing. There's no Chinatown, Little Italy, or any of that. The culture in Worcester is virtually non-existent unless you're in college, and in a city with as many colleges as Worcester, you'd think something more than a Denny's would be open past 1am. (I can't even find a damned diner.)
Hartford's not much different. At least it's got the Colt Armory and a few other hidden gems, but even those aren't all that spectacular. New Haven's the closest thing to a cultural center between Boston and New York, and that's only because of Yale and the Peabody Museum. Newport's got some nice historical stuff, but that's it.
But honestly, what exactly do you define as culture? Something that you subject yourself to everytime your significant other wants to broaden their minds? Or is it what the average person does for fun that dictates what a cultural experience is? I tend towards the latter for a more accurate description, and let me tell you something about that. If you aren't in a major city, then culture consists of hanging out at all-night diners, catching the occasional movie, and watching TV.
Worcester. Hell, last time I was in Portland Maine, there was more to do than there is in that pit of entropy known as Worcester. Biggest cultural event here is dodging potholes, and the chairs that people use to hold their parking spaces.
Sorry if this rambles, but I'm looking out my window and wondering why I moved here. At least the rent's cheap.
Sure, there's a Linux driver and all that jazz, and it works fairly well, etc. etc., but M Systems has a reputation in the Embedded Linux community for wanting you to sign your soul over before being able to use their stuff. The license is HIGHLY proprietary on their Disk-On-Chip product, and I'm sure that something similar will hold true for their other products.
If you aren't a zealot when it comes to licensing, then you've got a pretty decent USB storage device here. If you're a stickler for non-proprietary systems, then avoid it like the plague. The Linux Router Project and the LEAF Project have been looking for a way to incorporate something larger than the floppy for embedded firewall/router boxes, and M Systems' stuff was looked at and discarded due to the licensing issue.
Stuff to be aware of. Why not get yourself a cheap IDE-based Compact-Flash card reader if you need the space? They've got them for around $20 plus flash, and with flash being used in digital cameras, you can get some pretty decent-sized ones for cheap.
There's a distinct difference between charging "Exorbitant" fees and screwing your client over by telling them to purchase something that they don't want or need. Best example is the story (Urban Legend?) of the consultant who got called in to fix some constantly crashing servers. Took him 5 minutes, and he charged a grand to do it. When asked for an itemized bill, it said:
Fix Crashing Servers: $5
Knowing How to Fix Crashing Servers: $995
Supposedly, the company paid. Why? Each crash cost them an hour of productivity. Say it happened three times a week and affected 50 employees being paid $30,000 a year. That comes out to roughly $14.00 an hour each. Multiplied by 3 times a week and 50 users. Total cost per week to keep the problem, $2100.00. What you or I call exorbitant is pocketchange to a company facing a productivity loss on the order of $2100 per week.
This is totally different than, say, someone who needs remote access to all of their locations for routine maintenance and occasional file transfers, yet the consultant - rather than going with Frame Relay or possibly even DSL - tells the company that the only way to handle that is to go via T-3 to each major site, and load-balanced T-1 pairs to the minor ones. Each site's total traffic at peak times won't exceed say 400kbit/sec, but this consultant just sold them on 45Mbit/sec worth of bandwidth. It's unethical, it's wrong, and he's costing the company tens of thousands of dollars each MONTH that they don't need to be paying, just so he can get a bigger check.
It's not the fees that consultants charge, it's those of us who sell someone something that they legitimately do not need and will cost them MUCH more than just the consultant's inflated fee. I call it inflated by the way, because if they're doing such things, it's to get the bigger payout.
I had a feeling that this might actually be the case, but I didn't think it would make a difference of more than a minute and a half on a single-thread compile. Especially with a slower processor, slower FSB, and slower memory. The FSB and memory should actually help make up for the difference.
I'm also still puzzled by the randomness of when the compiles would run incredibly fast. It was done after removing and re-extracting the directory, in separate directories, etc.
Will the B+ Tree method used by Reiser call previously-used files that are identical, but in different directories, for use in functionally identical processes? Anyone?
Being bored and with a comparable machine, I decided to do some tests of my own.
System: SuSE 7.0, kernel 2.4.1 compiled with Uniprocessor and APIC/IO_APIC.
Athlon 1.1GHz, Asus A7V motherboard. FSB is 100MHz DDR. Memory is 256 megs at PC133, ATA66 5400RPM drive with ReiserFS.
I performed three series of tests. All tests were performed in single/double/triple thread orders, and each thread compile had it's own directory.
First test, all three had been make config'd per the original article, followed by make dep. After that, I rebooted and did all three compiles without rebooting. Second series started the process over again by make mrproper/make oldconfig/make dep/time make -jN bzImage, with N being the corresponding thread. Finally, I did a make mrproper/make oldconfig/make dep and rebooted each time before the compile.
I should note that on several occasions, I got Odd results; whether this was caching of some sort or not I don't know, but I would get 3m35s on a single thread and 1m9sec on a -j2 with a removed and recreated directory, as well as one or two other occasions - unfortunately, all the other occasions were when I was accidentally failing to use "time make -j2 bzImage" and instead was only doing "make -j2 bzImage", so I have no empirical proof. At any rate, here's the recorded ones.
Round 1
Straight
real 3m17.571s
user 2m54.660s
sys 0m13.120s
-j2
real 3m13.772s
user 2m58.390s
sys 0m13.390s
-j3
real 3m13.470s
user 2m59.390s
sys 0m13.180s
Round 2
Straight
real 3m8.048s
user 2m54.780s
sys 0m13.140s
-j2
real 3m11.912s
user 2m58.050s
sys 0m13.590s
-j3
real 3m12.532s
user 2m58.370s
sys 0m13.900s
Fresh-boot compile
Single thread was not redone; it was the Round 1.
-j2
real 3m15.634s
user 2m58.030s
sys 0m13.700s
-j3
real 3m16.433s
user 2m59.310s
sys 0m13.290s
As you can see, not much of a variation on here. The times are also a hell of a lot better than a 1.2GHz system single-threaded with DDR SDRAM, which makes me wonder what precisely is slowing down the 1.2GHz...
I work for an ISP with a rather large self-owned backbone - OC-48 rings all over the East Coast with OC-3/DS-3 supplementing; and multiple OC-3 cross-country lines for connectivity to Chicago and San Francisco - and the process is simple. We get an e-mail about Spam into abuse.
Abuse contacts the end user and tells them to fix the open relay within X amount of time. After the expiration of X, if another e-mail comes in, abuse investigates, uses... somewhat more firm language... with the end user. They then have one half X amount of time. After THAT, they contact my group - Operations - and say, sink the route. We go in, route the IP of the offending open relay to Null0, and wait for the guy to call. Then we direct him to abuse. Until such time as Abuse is sure that the relay is closed, the IP stays routed to Null0. When they're happy, we route it back to their circuit.
Spammers, on the other hand, go through step one, and on step two, they get shut down by us. As in, we disable the entire circuit, not just their IPs.
As for MAPS contacting the owners, they do in fact do just that. We somewhat regularly get e-mails from MAPS about it on the Operations mailing list. When they say they notify the owner, they mean the ARIN-Registered owner of the IP block, not the owner of the individual sites. Blacklisting a Spam-friendly ISP would be significantly more effective in getting said ISP to play ball by contacting innocent site owners and making them aware of this.
That being said, we don't use MAPS, and the fact that they're punishing "innocent" - not-guilty if you prefer - website owners is totally out of line and doesn't help much. These owners most likely don't realize that they've been RBL'd, and will still pay their bills to the ISP that's treated them well. Getting them to move is the objective; not telling them that they may want to is defeating that objective.
Personally though, I don't find it all that difficult. Why? Because I go to SuSE's ftp site, poke around, and lo and behold, I find a directory with the INSTALL DISKS FROM THE CD DISTRO.
Sheesh. Can you really blame them for wanting you to buy instead of download? Do you think they make their money on the free support that comes with a purchased distro? No, they make it on the damned fine manuals and - in the case of SuSE 6.4 - the mini-FAQ and basic install instructions printed on the sleeve.
Simple fact is, they're a BUSINESS. They don't need to make it EASY for you, but they DO make it available. And if you've really got issues where you can't download and dd a floppy image to a disk after spending some time poking around a directory tree, then please don't complain that they're not giving away their lifeblood.
After all, RMS charged an arm and a leg for a hardcopy of Emacs. I fail to see where a full distro should do any differently. I'm not big on the Personal/Proffessional distinction, but RedHat does the same thing. Nothing new here.
Wolfstar - waiting for the ftp site to un-Slashdot itself before upgrading.
I've used SuSE 6.2 and 6.4, and since switching to 6.4, I've been running ReiserFS(Sometimes nuking your box can be a good thing.) under the following kernels:
2.2.14-SuSE, 2.2.16-SuSE, 2.3.51-SuSE, 2.4.0-test6 through test8.
Not only have I not had a problem with hdparm, I've managed to hard-off my box - thank the cats playing with power cords - at LEAST once per kernel variant with no loss of data or problems. ReiserFS wasn't a volatile inclusion on SuSE's part; they're a primary sponsor of it as well as sponsoring XFree86 development.
To be honest, the only thing I can think of is that it was a once-off, because I haven't seen any problems with Reiser, and it's saved me a lot of heartache.
I read the patent application. I read the posted comments. I noted that this has been in effect for TWO YEARS NOW without a worry. And I spend 10 hours a day, 5 days a week inside a Cisco router.
The only thing that this patent is doing is allowing Static NAT with NAT pools OR one-to-many NAT a la masquerading to be used without compromising the effectiveness of a firewall; most likely in this case Cisco's Access List filters. What they're doing is patenting a method of applying filters based on internal network addresses from external hosts and not blowing A) System Integrity or B) Efficiency out of the water. And what it does it does very very well. The basic/standard Linux firewall and routing routines currently released - no, I'm not talking about the 2.4.0-test series - can only just barely keep up with what a Cisco 2500 with 4 megs of ram can do with a pair of T-1s and a large network behind it.
Believe me. If we haven't seen it yet, we're not going to. That's because they're NOT PATENTING NAT.
Actually, Covad's not booked on weekends. The exact opposite. Their Technical Assistance Center has one person in on weekends in the dispatch department to handle incoming requests for dispatch and schedule them for the next week. Covad's Dispatch techs don't work weekends, and only rarely stay after 5pm.
Slashdot Mirror
First off, he says that out of the 3500 students invading the campus, 1500 of them will be screaming Mommy when they head in and try and get connected. This is about right for A) The incoming Freshmen, and B) The terminally stupid upperclassmen.
Also, the number of people bringing computers to school with them and thinking that the archaic 8088 XT that they just dug out of the basement - usually because their parents can't or won't let them take the high-end 486 that the family uses - might be a bit surprising. (This is of course an exaggeration, I hope. None of my friends who've been there and done that ever mentioned anything quite so drastic.)
Also, there's the fact that, while the NETWORK might be able to take the abuse, it's not guaranteed that the Network ADMINS can handle the stupidity. Super-cheap-laptop + Win2K + Novell + Wireless = Twitching Admin. I really can begin to imagine the hell of it all compressed into three days or so, because - even if it wasn't tech related - I've worked the bookstore during hell week at a fairly large University before. You can't begin to imagine the disruption of life that occurs to the people who work on campuses at the end of the summer unless you've been one.
And yes, I imagine he IS in fact crying over his lost phat pipe. =)
For starters, the PEERING arrangement is gone. There's a difference between peering and transit. Peering is done when it would be mutually beneficial for both parties to share traffic, and some arrangement regarding the cost of the line - in the case of a private peer such as this - is worked out between the two. Transit is where the ISP pays out the nose for a link to the Network provider's network, allowing the ISP to use the Network provider as an upstream to access most of the world.
Now, I work for a fairly large ISP, with a nationwide privately-owned fiber backbone. We get 16ms ping times from Maine to the Carolinas, and 70-80ms pingtimes from New York to San Francisco. And we've got TONS of peering arrangements. But there's three networks that we still aren't big enough to get transit with: Level3, UUNet, and Cable & Wireless.
Big network providers like C&W don't peer with the small fries, and if you have a peering arrangement with one of them, you're right near the top - Tier One or Two Network provider. If C&W is dropping peering for PSINet, that means that, in their opinion, PSINet isn't classed as a Top-Tier network provider anymore. And that is the sound of the bell tolling doom for any network provider out there. That's why this is news.
This is JAPAN. The culture CANNOT GET MUCH DIFFERENT than the United States.
Corporate Culture in Japan is something totally different than going to the bar after work with a few coworkers, and is much more akin to citizenship in a very patriotic country. Health Club, medical, dental, living spaces, restaurants, movie theaters, all of this can and in some cases usually is provided by the company for their employees. Read up on the RPG Shadowrun to get an example of Japanese Corporate Culture taken to an extreme, then realize it isn't too extreme.
Furthermore, the corporate world of Japan is very much their replacement for the warriors of old. Employment with a corporation borders on feudal vassalship, and this is not only accepted, but considered normal by many. Sitting over here stateside, where an employer wouldn't dare try a stunt like that, and saying that it should be illegal in a society that doesn't match ours is a little bit ridiculous.
Your Mileage May Vary, but that's the reason *I* see for these particular actions, and I don't see what the big deal is.
Also, if these have been turning up since the 1800s, is there a major archaeological relevance to that area? What was there at the time, anyone know?
The point that he was making here was fairly simple. It was something that was pointed out on CNN a little over a year ago. The day that the RIAA announced their lawsuit against Napster, the blurb that followed it was "on a related note, CD sales are up an estimated x% this quarter" - paraphrased, and I think x was around 15 or so. There was a demonstrated increase in CD sales once Napster became popular. Whether or not it's due to Napster is another story, but I think it is.
All governing bodies have some amount of evil, and it's easy to overlook the good when all you care about is what they are depriving you of (free music).
Statement is true, concept is ... flawed in my opinion. If artists were making more than pennies on the dollar for their music, I might be able to accept this as a valid argument. But this is the Recording Industry Association of America, not the Recording Artist's Association of America. RIAA exists to protect the recording studios primarily, and as a side effect, it sort of protects the companies' artists - from everyone except the companies themselves, that is.
What makes you think that Micorosoft should give away upgrades to their software, simply because you personally gauge the price to be too high? I know it's a fair whack, but to think that you are getting all that product (consider the developer's time that went in to making this stuff) and you just think you are welcome to free upgrades?
Yes.
Reasons why: I go out and buy Adaptec EZ CD Creator 4.0 so that I can burn backups of all my text files - and I have a ton of them. I go home, find out that 4.0 doesn't support my neato-nifty-keen 16x10x40 burner that I got white-boxed from some 'net retailer. I hit the EZ CD Creator Website, and I see that there's an update to version 4.02, install that, it works with my burner. I'm happy again.
Now, you're most likely to say, "Yeah, but that's a point-release, not the same." My response is, tell me one thing about Windows 98 that couldn't have been done with a point-release type of update. I really don't see it. Let's say I upgrade my old Pentium 233 to a K6-2 500, and I'm running Windows 95. Suddenly, my computer won't finish starting Windows, and it's a SOFTWARE issue. I have two choices, fall back to my Pentium 233 and hope that I can find the executable - only usable in the Windows interface, by the way, which means you HAVE to fallback or underclock - that is buried somewhere in Microsoft's website, or go out and drop a hundred bucks or more for Win 98. This is a problem I've run into recently - also affects Athlons in excess of 1GHz under Win95 - but it's not included in Windows Update, and it's so well hidden that it took me two hours to find it. This is a software bug on Microsoft's part. Why isn't it included in Windows Update?
Microsoft actually bucks industry standard practice when it comes to updates. It should cost significantly less to upgrade if there's no way to add compatibility for new hardware to the old OS. I do agree, however, that the comment about the WPA, while not two-faced per se, is a contradiction in concepts.
Everyone's entitled to an opinion I guess, but this is clearly just a college kid that's pissed he doesn't get enough pocket money.
This I think is a bit out of line, since he's really not griping about not having enough money. He's griping about software prices.
Example: I'm a network engineer. I make a fairly good living, drive a sporty type of car made in 1997, have plenty of money left over to eat with once I pay all the bills, etc. Overall, I'm at a position where I'm living comfortably. Why is it that if I wanted to go out and buy, say, Microsoft Word or Excel, or heavens forbid Visual Studio, that I would have to save up for the next six months to buy it - either that or skip the rent AND the car payment - just so I can have a decent, industry-standard word processor and spreadsheet? Some will say get StarOffice, or maybe suggest AbiWord under Linux. That doesn't work too well. You go looking for a new job, and every recruiter or headhunter you come across is going to tell you to send them a resume in Word format. You ever try writing a resume in wordpad? Not the easiest thing to do and still have it presentable.
This is also moot. Most software companies that make productivity software simply grit their teeth and ignore the individuals with pirate copies - for the most part. It's the businesses with site licenses that they need to worry about. One company is literally hundreds and thousands of dollars in revenue.
It may not be news, but then again, neither are a lot of Op-Ed pieces in a traditional newspaper. They still see press though. Why shouldn't this?
Stormix crashed and burned on my system the three times I tried installing it, and I didn't have anything too obscure in there either. Potato and Slink are both fairly awkward to install, but you can usually get to a shell prompt easily enough.
However, I personally take a wee bit of umbrage at the "easy way to install Debian" bit in the main article, because it is most definitely NOT the only way to do it. And Slashdot is what put me onto the alternative in the first place.
In my opinion, the Holy Grail of Linux Distros, the one that does what even Mandrake (7.2, I haven't been able to get 8.0 yet) can't do for ease of install, is Progeny Debian GNU/Linux.
Note that I don't use this on a day-to-day basis, mostly because of the blood-sweat-and-tears I've put into my SuSE boxen, but this is THE single easiest-to-install distribution out there that I've come across. It pegged ALL of my hardware - even my Logitech Cordless trackball, which SuSE's quite excellent SaX and SaX2 programs could not - first go around, and was easier to install than Windows.
Yes. Easier to install than Windows. Not for us, the average geek, but for them, the average Windows user.
Now, there are a few problems/bugs with the Progeny that I played with - namely, minor bugs in the installs package selection program that causes things to go wonky if you unselect a package in Expert Install - but overall, it's smooth as hell. One of my friends at work tried it after I suggested it to him; he installed in in about half an hour, got it up and running, and grabbed Ximian Gnome, and he's happier than hell. It's based off of a snapshot of Woody, so whether the installer is Woody's or Progeny's I don't know. What I do know is, if I ever need to do a system reinstall, SuSE will only remain on my server, and Progeny will be my desktop. SuSE's been good to me, but Progeny is unbelievable.
Before you wonder, no, I don't work for Progeny. But there's quite a few Debian notables that do.
SDSL is far more important than DSL overall, yes. Unfortunately, it's significantly less stable. As soon as they can get some more robustness into things - like, not having to worry about bad rainstorms (I've lost track of the number of DSL circuits that have gone down due to wires getting wet) then I'll be recommending SDSL to everyone.
Honestly though? I've got an ADSL line at home, and it's 1.5 down and 384 up. A lot of people don't realize that you can run a decent number of services over 384k and not have to worry about the bandwidth. It also will leave you room on the downstream for browsing, even if your server is getting smacked. So I honestly can't say that SDSL lines are necessary for services.
Thanks for the comments.
I work in NetOps myself, and I hadn't heard of a T-6 until about two months ago. Verizon uses them, at least in New England, for minor backhaul carrier circuits. From what I understand, a T-6/DS-6 is comprised of 24 DS-3 channels. Talk about your fat pipe.
To the best of my knowledge, they're rarely used anymore, and are really only leftovers from areas that haven't been fully rolled to fibre backhauls. Then again, this is Massachussetts, and we've still got a few places with pulse-dial only switches. Go figure.
This is not correct. Yes, I'm being blunt.
The REASON this isn't correct is because of the design of a DSLAM. DSLAMs are made up of card shelves. Each card has X number of customers - I think that Covad SDSL cards allow 6, but don't hold me to that - and there are multiple cards per shelf. There's at LEAST one ATM DS-3 running off of each DSLAM into the Covad networks - Covad is what I have the experience with - and they add more pipe based on the number of users.
This CAN lead to problems, especially if 30 or so users decide to use their full bandwidth all of the time. DSL ISPs have ways of handling this, usually by warning them to stop doing it 24/7, or by shutting them down. Either way, with ATM circuits, you can have some five or six hundred people going over a single DS-3, and still only use about 20Mbit/sec worth of bandwidth on it.
As for the phone pair, it does NOT need to be up to date and in top condition; Harvard.net - one of the first DSL providers to go under, and I think exclusive to the Northeast - was buying "dry" pairs with no conditioning done to it from Verizon. DSL will run over crufty old alarm pair for fire alarms and the like; just don't expect it to stay up and running all the time. A lot of the DSL providers out there are now getting conditioned pairs to use for lines, but it's by no means necessary.
As for your brother's download speeds, that depends entirely on the ISP. If they've got an overloaded backhaul, or their transit to upstream providers is full, then peak times are going to suck. LOTS.
One person I know of is currently working on his Battlebot right now, and the basis for it is going to be a heavily modified Linux Router Project disk, using CompactFlash for the "disk", one of the Linux BIOS projects for boot, and a couple of wireless LAN cards for controls. He's keeping a radio in it for backup, but I'm betting that Battlebots won't ever be the same again.
Not to mention the fact that there's a lot more tech in the R/C and Ham Radio worlds than most purely computer geeks want to give them credit for.
For starters, there's the old saw about the shared node. With DSL, you don't need to wonder what's going to happen to your bandwidth when the rest of the neighborhood wakes up and smells the fat pipe. Another big one is, 99% of the cable providers out there throw a fairly big stink when you try and run a server over a cablemodem - once they get around to noticing it. My DSL ISP - Speakeasy - has two restrictions: No porn websites, and no IRC Servers. (And even that they can overlook if it's a single-node server and not part of a network, I think.)
In addition to that, DSL is cheaper. How? What if I don't WANT cable TV? Most cable companies won't sell you cablemodem access without a TV subscription as well. (Rare, but true. One of my friends is stuck with this dilemma.) Also, DSL is available where many times cablemodem isn't. Reverse is also true, of course, but unlike cablemodem access, you don't need to upgrade the existing infrastructure to accomodate DSL. Just stick a DSLAM in a central office and start plugging in lines.
Lastly, very few cable companies want to give you the same bandwidth upstream as you have downstream, without paying a LOT of money for it. DSL is comparatively cheap for synchronous speeds.
Hope that answers some of your questions. Incidentally, line conditions for DSL and Cable both are much better than average in Seattle, since there isn't a couple hundred years of cruft in the way.
Your business class DSL comes with no uptime guarantees. Telcos purposely drag their feet when they have to dispatch on a DSL. Support varies from ISP to ISP badly enough. And worst of all, there is NOTHING ELSE with a lower priority than DSL when there's major storms/natural disaster outages. Priority goes like this:
1. Optical Backbone circuits - OC-192 to OC-3
2. T-6/T-3 lines
3. 911 Emergency Service
4. T-1 lines
5. Frame Relay
6. ISDN Lines
7. Voice/POTS lines
8. DSL
Further, you will get LESS bang for your buck at equivalent speeds, because - simply put - you've got a 10% overhead built-in, since most DSL Providers use ATM-based DSL setups. 5 out of every 53 bytes of data is header information that gets stripped out by the modem or router.
If you've got a business, and you need an internet connection, get a T-1. Or even a Frame Relay. Anything else and you might as well be gambling.
This isn't necessarily true. The local telcos are legally only able to charge a minimal fee to CLECs - Competitive Local Exchange Carriers, which is technically what a DSL provider is - of around $7 for the local loop. The DSL providers then turn around and sell access to their network for X number of dollars, plus a fee per circuit of - I THINK - around $40 per line. (This is what it was about a year ago; whether or not that's changed since then, I don't know.) On top of that, you get whatever else the ISP charges. The rates aren't cut that much really; you're looking at an over 300% profit on the loop charge alone, and I know that most DSL ISPs that are worth looking at twice are charging around $60 for a 608/128 line, $90 for 1536/384 lines, and around $50 for a 192k SDSL line, working up. (Highest I've seen is $450/month for a 1.5Mbit SDSL line with some real nice side perks.)
DSL is far from cut-rate; the companies involved just happened to jump in too early to gain the early ground on it. Support costs a lot for DSL, although that's changing with the new Rate Adaptive Error Correction software out there. My line had a hard short on it at one point about two months ago, that I found out about quite by accident. My DSL line was still up through it, and - having troubleshot for umpteen million DSL circuits - I know that a hard short on a SDSL line takes it down. (In case you're wondering, the reason it isn't applied is because it would break the guarantees of speed on an SDSL line. The error correction drops your speed to compensate for loss of signal strength.)
I agree that DSL doesn't really work - currently - but I don't agree that it's because of cut-rate pricing. These companies are making up lost ground, but they're doing it slowly, because they can't afford to tinker with it to get it to work better.
Semi-useless factoid for the day. Copper-based T-1s are usually HDSL, which is the oldest of the DSL types. It's the repeaters and switching equipment - along with the cost of the loop and REALLY strict Service Level Agreements - that make all the difference in the world.
The only purges that Covad has been doing are those where they're pulling gear out of thoroughly unprofitable COs - because, for instance, there aren't enough subscribers to afford the cage (we're talking two or three people in the CO here) - and the shutdown of service to those ISPs that are unwilling or unable to pay their bills to Covad. If anything, this made them MORE reliable as a long-term prospect. They dropped their dead weight, and their customer service at it's worst was light-years better than Northpoint's average - and still a bit better than Northpoint at it's best.
If someone got caught in a purge, then they went with a dinky little ISP that didn't pay their bills and didn't let it be known that they were in danger of losing connectivity. These providers - ALL providers that sold Covad DSL for that matter - were notified of the impending shutdowns. Covad even had a project set up to help customers transfer to new ISPs. Speakeasy was one of them - and it's still going on. (Semi-Disclaimer: I work for RCN, but Speakeasy is my DSL Provider. There's reasons, but most of it boils down to cost.)
Rule Number One of DSL: If your business relies on it's net connection, DO NOT UNDER ANY CIRCUMSTANCES USE DSL AS A PRIMARY CONNECTION! I cannot stress that enough! DSL has absolutely NO service agreements, beyond a guaranteed 80% of rated line-speed for "business" class SDSL circuits. DSL is great for the home, and good for a backup circuit, but don't think it's a T-1 for cheap.
Trust in the fact that, if ANYONE goes ANYWHERE anytime soon, it won't be Covad. They're running strong, they laid off a few people back about 6-8 months ago. After that, their support IMPROVED. They reorganized, and they're running better than ever. And their service is better, more reliable, and usually more intelligently-staffed than just about any of the local Telcos are. Do your homework on DSL, and you won't get burned.
I do commercial Level 2 Support for a large national ISP. We had a bunch of customers through Northpoint, and the network went down hard yesterday. My personal view on why they went under is that their support was horrendous.
When your line went down - and if you have DSL, your line WILL go down someday - you call into your ISP. Your ISP verifies their side, and then called Northpoint to open a ticket. You called into Tier 1, which basically took calls, noted down information, and read back to you what was posted on their support website. This info was largely useless, because as recently as October, right around when the deal with Verizon first came up, Northpoint used a paper-only trouble ticket system. Yes. Paper Only.
If, after two or three hours, the ISP hasn't heard anything or been contacted for testing, with Covad, you call up and talk to someone who can do the testing. With Northpoint, you had no choice but to call the Tier 1 folks, who promptly told you that there was nothing that could be done because they have four hours to respond. (I once called 2 hours in, sat on hold for an hour and 45 minutes for Tier 1, and when they picked up, told me that there was nothing that could be done, then hung up on me.) No matter how far you escalated the issue within their management, no matter how long the ticket had been open with them, you could not call any sooner than four hours and expect anything to get done.
Eventually, once the deal with Verizon fell through, this began to change. Rumor has it that they started using a computerized ticketing system at some point in there, and they removed the four-hour requirement. We were provided numbers that went directly to the Tier 2 testers, and sometimes things actually got done. There were still annoyances - if a customer went down, and it was determined that their router or DSL Modem went bad, a tech was sent out, tested with a spare modem, verified it, and then took the spare modem with them when they left!!!! Issues such as this were the norm, and why Northpoint went under is that their Partner ISPs - like us, who had over a thousand DSL customers through them - were constantly given the runaround when it came to support, and they didn't realize this until they were right on the brink, with Verizon backing out and virtually no hope of actually securing investment capital.
Northpoint had bad business practices and poor support, realized it way too late to do anything about it, and paid the price. Maybe their service was cheaper and more reliable than Verizon. But don't forget that you get what you pay for, and don't make the mistake of thinking that they were a decent company.
The entire historical significance of Worcester Massachussetts is that Robert Goddard popped a rocket off a farm somewhere near it. In the past year and a half, I have yet to see anything else worth mentioning.
As for the culture, I live right in the middle of a heavy Italian neighborhood, and there's a place down the street where I can get handmade Italian sausages for the sauce that I make every so often. Some Thai/Polynesian places, some assorted other cultural backgrounds here and there, but by and large, there's nothing. There's no Chinatown, Little Italy, or any of that. The culture in Worcester is virtually non-existent unless you're in college, and in a city with as many colleges as Worcester, you'd think something more than a Denny's would be open past 1am. (I can't even find a damned diner.)
Hartford's not much different. At least it's got the Colt Armory and a few other hidden gems, but even those aren't all that spectacular. New Haven's the closest thing to a cultural center between Boston and New York, and that's only because of Yale and the Peabody Museum. Newport's got some nice historical stuff, but that's it.
But honestly, what exactly do you define as culture? Something that you subject yourself to everytime your significant other wants to broaden their minds? Or is it what the average person does for fun that dictates what a cultural experience is? I tend towards the latter for a more accurate description, and let me tell you something about that. If you aren't in a major city, then culture consists of hanging out at all-night diners, catching the occasional movie, and watching TV.
Worcester. Hell, last time I was in Portland Maine, there was more to do than there is in that pit of entropy known as Worcester. Biggest cultural event here is dodging potholes, and the chairs that people use to hold their parking spaces.
Sorry if this rambles, but I'm looking out my window and wondering why I moved here. At least the rent's cheap.
Sure, there's a Linux driver and all that jazz, and it works fairly well, etc. etc., but M Systems has a reputation in the Embedded Linux community for wanting you to sign your soul over before being able to use their stuff. The license is HIGHLY proprietary on their Disk-On-Chip product, and I'm sure that something similar will hold true for their other products.
If you aren't a zealot when it comes to licensing, then you've got a pretty decent USB storage device here. If you're a stickler for non-proprietary systems, then avoid it like the plague. The Linux Router Project and the LEAF Project have been looking for a way to incorporate something larger than the floppy for embedded firewall/router boxes, and M Systems' stuff was looked at and discarded due to the licensing issue.
Stuff to be aware of. Why not get yourself a cheap IDE-based Compact-Flash card reader if you need the space? They've got them for around $20 plus flash, and with flash being used in digital cameras, you can get some pretty decent-sized ones for cheap.
Fix Crashing Servers: $5
Knowing How to Fix Crashing Servers: $995
Supposedly, the company paid. Why? Each crash cost them an hour of productivity. Say it happened three times a week and affected 50 employees being paid $30,000 a year. That comes out to roughly $14.00 an hour each. Multiplied by 3 times a week and 50 users. Total cost per week to keep the problem, $2100.00. What you or I call exorbitant is pocketchange to a company facing a productivity loss on the order of $2100 per week.
This is totally different than, say, someone who needs remote access to all of their locations for routine maintenance and occasional file transfers, yet the consultant - rather than going with Frame Relay or possibly even DSL - tells the company that the only way to handle that is to go via T-3 to each major site, and load-balanced T-1 pairs to the minor ones. Each site's total traffic at peak times won't exceed say 400kbit/sec, but this consultant just sold them on 45Mbit/sec worth of bandwidth. It's unethical, it's wrong, and he's costing the company tens of thousands of dollars each MONTH that they don't need to be paying, just so he can get a bigger check.
It's not the fees that consultants charge, it's those of us who sell someone something that they legitimately do not need and will cost them MUCH more than just the consultant's inflated fee. I call it inflated by the way, because if they're doing such things, it's to get the bigger payout.
I'm also still puzzled by the randomness of when the compiles would run incredibly fast. It was done after removing and re-extracting the directory, in separate directories, etc.
Will the B+ Tree method used by Reiser call previously-used files that are identical, but in different directories, for use in functionally identical processes? Anyone?
System: SuSE 7.0, kernel 2.4.1 compiled with Uniprocessor and APIC/IO_APIC.
Athlon 1.1GHz, Asus A7V motherboard. FSB is 100MHz DDR. Memory is 256 megs at PC133, ATA66 5400RPM drive with ReiserFS.
I performed three series of tests. All tests were performed in single/double/triple thread orders, and each thread compile had it's own directory.
First test, all three had been make config'd per the original article, followed by make dep. After that, I rebooted and did all three compiles without rebooting. Second series started the process over again by make mrproper/make oldconfig/make dep/time make -jN bzImage, with N being the corresponding thread. Finally, I did a make mrproper/make oldconfig/make dep and rebooted each time before the compile.
I should note that on several occasions, I got Odd results; whether this was caching of some sort or not I don't know, but I would get 3m35s on a single thread and 1m9sec on a -j2 with a removed and recreated directory, as well as one or two other occasions - unfortunately, all the other occasions were when I was accidentally failing to use "time make -j2 bzImage" and instead was only doing "make -j2 bzImage", so I have no empirical proof. At any rate, here's the recorded ones.
Round 1
Straight
real 3m17.571s
user 2m54.660s
sys 0m13.120s
-j2
real 3m13.772s
user 2m58.390s
sys 0m13.390s
-j3
real 3m13.470s
user 2m59.390s
sys 0m13.180s
Round 2
Straight
real 3m8.048s
user 2m54.780s
sys 0m13.140s
-j2
real 3m11.912s
user 2m58.050s
sys 0m13.590s
-j3
real 3m12.532s
user 2m58.370s
sys 0m13.900s
Fresh-boot compile
Single thread was not redone; it was the Round 1.
-j2
real 3m15.634s
user 2m58.030s
sys 0m13.700s
-j3
real 3m16.433s
user 2m59.310s
sys 0m13.290s
As you can see, not much of a variation on here. The times are also a hell of a lot better than a 1.2GHz system single-threaded with DDR SDRAM, which makes me wonder what precisely is slowing down the 1.2GHz...
Food for thought.
I work for an ISP with a rather large self-owned backbone - OC-48 rings all over the East Coast with OC-3/DS-3 supplementing; and multiple OC-3 cross-country lines for connectivity to Chicago and San Francisco - and the process is simple. We get an e-mail about Spam into abuse.
Abuse contacts the end user and tells them to fix the open relay within X amount of time. After the expiration of X, if another e-mail comes in, abuse investigates, uses ... somewhat more firm language ... with the end user. They then have one half X amount of time. After THAT, they contact my group - Operations - and say, sink the route. We go in, route the IP of the offending open relay to Null0, and wait for the guy to call. Then we direct him to abuse. Until such time as Abuse is sure that the relay is closed, the IP stays routed to Null0. When they're happy, we route it back to their circuit.
Spammers, on the other hand, go through step one, and on step two, they get shut down by us. As in, we disable the entire circuit, not just their IPs.
As for MAPS contacting the owners, they do in fact do just that. We somewhat regularly get e-mails from MAPS about it on the Operations mailing list. When they say they notify the owner, they mean the ARIN-Registered owner of the IP block, not the owner of the individual sites. Blacklisting a Spam-friendly ISP would be significantly more effective in getting said ISP to play ball by contacting innocent site owners and making them aware of this.
That being said, we don't use MAPS, and the fact that they're punishing "innocent" - not-guilty if you prefer - website owners is totally out of line and doesn't help much. These owners most likely don't realize that they've been RBL'd, and will still pay their bills to the ISP that's treated them well. Getting them to move is the objective; not telling them that they may want to is defeating that objective.
Wolfstar
Well then. I guess we know where you stand.
Personally though, I don't find it all that difficult. Why? Because I go to SuSE's ftp site, poke around, and lo and behold, I find a directory with the INSTALL DISKS FROM THE CD DISTRO.
Sheesh. Can you really blame them for wanting you to buy instead of download? Do you think they make their money on the free support that comes with a purchased distro? No, they make it on the damned fine manuals and - in the case of SuSE 6.4 - the mini-FAQ and basic install instructions printed on the sleeve.
Simple fact is, they're a BUSINESS. They don't need to make it EASY for you, but they DO make it available. And if you've really got issues where you can't download and dd a floppy image to a disk after spending some time poking around a directory tree, then please don't complain that they're not giving away their lifeblood.
After all, RMS charged an arm and a leg for a hardcopy of Emacs. I fail to see where a full distro should do any differently. I'm not big on the Personal/Proffessional distinction, but RedHat does the same thing. Nothing new here.
Wolfstar - waiting for the ftp site to un-Slashdot itself before upgrading.
That is REALLY odd.
I've used SuSE 6.2 and 6.4, and since switching to 6.4, I've been running ReiserFS(Sometimes nuking your box can be a good thing.) under the following kernels:
2.2.14-SuSE, 2.2.16-SuSE, 2.3.51-SuSE, 2.4.0-test6 through test8.
Not only have I not had a problem with hdparm, I've managed to hard-off my box - thank the cats playing with power cords - at LEAST once per kernel variant with no loss of data or problems. ReiserFS wasn't a volatile inclusion on SuSE's part; they're a primary sponsor of it as well as sponsoring XFree86 development.
To be honest, the only thing I can think of is that it was a once-off, because I haven't seen any problems with Reiser, and it's saved me a lot of heartache.
Wolfstar
Okay.
I read the patent application. I read the posted comments. I noted that this has been in effect for TWO YEARS NOW without a worry. And I spend 10 hours a day, 5 days a week inside a Cisco router.
The only thing that this patent is doing is allowing Static NAT with NAT pools OR one-to-many NAT a la masquerading to be used without compromising the effectiveness of a firewall; most likely in this case Cisco's Access List filters. What they're doing is patenting a method of applying filters based on internal network addresses from external hosts and not blowing A) System Integrity or B) Efficiency out of the water. And what it does it does very very well. The basic/standard Linux firewall and routing routines currently released - no, I'm not talking about the 2.4.0-test series - can only just barely keep up with what a Cisco 2500 with 4 megs of ram can do with a pair of T-1s and a large network behind it.
Believe me. If we haven't seen it yet, we're not going to. That's because they're NOT PATENTING NAT.
Actually, Covad's not booked on weekends. The exact opposite. Their Technical Assistance Center has one person in on weekends in the dispatch department to handle incoming requests for dispatch and schedule them for the next week. Covad's Dispatch techs don't work weekends, and only rarely stay after 5pm.