I went through almost the same situation.. granted it was 10 years ago and at the time no one heard of something like a network engineer..
First off, I think a college degree can open a lot of doors. I'm fortunate enough to be doing exactly what I love in a part of the country most people can't afford to live in year round. I suspect I couldn't have done that with my resume alone. I can guarantee you won't get the hands on experience in an undergraduate program that someone getting a CCNA will get, but I don't think you should worry about that.
Now, getting down to some details. I think an Electrical Engineering, Computer Science, or Computer Engineering degree is what you should be looking at. I realized about my junior year I wanted to get into networking stuff and found I had some great resources to choose from:
Even though I was EE, I could take CE classes that counted as technical electives for my major. That let me take all the networking classes. (Ironically, it would have been harder to take them as a CE because they weren't allowed as many electives.)
I also discovered some graduate level seminars in networking. I took them all despite the fact I wasn't a grad student.
I got involved in an Aerospace project my senior year. Since I was the outsider I became the technical lead for communications/EE style stuff.
My first job was a sys admin. Excellent stuff and it gave me firm hands-on training. Now I do telecommunications, which mainly involves phone networks and infrastructure backbones for the computer geeks (fiber and copper in a campus environment.) I firmly believe you don't know shit about computer networks until you've worked on phone networks - it's kind of the missing piece that makes everything else make sense.
One other random thought.. someone told me once that college isn't about teaching you a profession it's about teaching you how to learn.
Actually, I haven't been a Linux sys admin for almost 4 years now. I did a nice 4 year stint admining Linux/Unix back when getting a job working with Linux was next to impossible and Knoppix didn't exist.
I still do that work on the side and still use Linux for personal use. But rather than having to admin 50 - 60 Linux servers I'm down to about 2. With a little luck I'll never have to do a rescue again. I've heard good things about Knoppix. Based on what you said maybe I should download it just to boot it up. Thanks!
I can't say I really care one way or the other about this.
Novell has been quite stagnant for a while. Thankfully they realized that they needed a shot in the arm in order to expand their increasingly legacy user base. But have they done anything interesting? Will they be a contributing member of the community? Why should I get excited about Novell?
Then you have SuSE. I've used the distribution quite a bit. More often than not for their Live CD to do rescues. It's an ok distribution - it definitely has more software than any other distro and the book they provide is excellent. But again, it's not that interesting of a distro. It's just enough of a pain in the ass to use to not be enjoyable.. unless you think Slackware combined with RPM's is cool or you enjoy finding random documentation written in German.
Can someone please tell me something I don't know that will make me care about two also-rans?
Um, I didn't really write anything about the reactions of Wine developers in this weeks' WWN. Save for agreeing with Mike Hearn that an announcement still seems pending.
Oh well, thanks for the press. It's always nice to see.
A lot of people have already mentioned the PhD-is-good-for-research theory, so I won't rant on that.
I've seen a lot of people lately who are staying in school rather than throwing themselves at a tough labor market. Personally, I think staying in school to escape "the real world" is a really weak excuse. If you've only been in school and don't have practical experience then I highly suggest getting out and getting a job. I don't get along as well with my friends who are still in school (I graduated five years ago with a BS). A lot of them seem to be in perpetual procrastination about putting together a resume, interviewing for jobs, and in constant pursuit of the next easy part-time job.
Now, having said that, perhaps you have some personal goals you'd like to accomplish or a certain area of study you really want to pursue. Doing that in a non-academic environment can be rough and then it makes sense to stay in school.
There's lots of exciting things to do in this world besides working on a thesis. While I think pursuing a career (you do have a grad degree) is good, I'm sure you could come up with something more original. Go work on a cruise ship for a year. Take off and lay on the beach in Thailand for a while. Spend the winter skiing.. of course that's what I did and accidentally never left.
If you're a tranport ISP and not on the edge I just don't see what options there are (were). Known exploits were in the wild and the only solutions are either Cisco's patch or something ACL based. ACL based might be fine if you can tolerate the increased load. Otherwise, the patch is it. And that sucks if you have to do
something like cross the 12.0(22)S hurdle.
Of course you should be careful, but if you're working in a large NOC and making the big bucks this is the type of thing you stay at work for hours testing. Yes, this exploit will wreck havoc for years, but hopefully not on any backbone provider's network. It's the time to stock up on caffeine and tell your wife you'll be home late.
With all the media attention most networks put out customer notifications of impending outage and I think most people thought it was ok.
As far as the Linux references (especially from the other guy), I don't buy it either. Until someone makes a nice canned Linux router product with Cisco style support it's simply not a solution. As far as Juniper goes, I think this is an excellent example of why it's a good thing to have mutiple router platforms in a datacenter. Of course I've said the exact opposite when it comes to having to support it..
Please don't assume a NOC is a large network scaled up. There are a million reasons why that's not the case. (sorry, couldn't resist)
There's no choice in the matter. This vulnerability has one fix available from Cisco (with the load raising ACL solution a distant 2nd) with known exploits in the wild. Very little extensive testing of this patch has been done and no one knows if the patch will break anything else. (Supposedly Cisco does 250,000 internal tests before releasing a patch.) Yes, there's unconfirmed reports of inout errors resulting from the patch.
If you are a transit ISP you test this as much as possible and roll it out. It's a game of Russian roulette and you're forced to play. If you're sitting in a mission critical NOC you apply this, deal with the fallout, and go to plan B in the event of something bad. That's why you're getting paid the big bucks, right?
If you're on the edge you probably have support to fall back on and in this instance it's exactly what you'd want to do.
Cisco released the fix two days ago to backbone providers. Other large customers could get the fix early yesterday. If you're affected by this vulnerability and it's not fixed yet:
You're not subscribed to the proper news channels (i.e. you're not doing your job) or
You're lazy (i.e. you're not doing your job) or
You're not as important as you thought (i.e. someone else isn't doing their job.)
It seems like Cisco handled this one correctly with the providers. I'm not sure how well large customers were handled, my guess is the.edu folks probably got screwed again.
We've been putting in some new hotel/condo buildings lately. Specifically we've built 8 new ones comprised of 500+ units. We made the decision to pull composite cable everywhere. We chose 2 RG-5, 2 Cat 5, and 2 strands of fiber in each cable. We used Siemen's "Home Cabling System" for all the terminations. Good stuff.
As for your existing building, recabling will be expensive. If you're lucky there's riser closets on each floor where you could use the existing cable to pull with. You could also locate hubs there. But that's a lot of work. The idea of using some low bandwidth solution to the unit isn't too bad. Hardware costs will be really high, but you may be able to pass it along to the homeowner rather than eat it yourself. Yeah, it'd be nice to have ethernet everywhere but that just might not be possible.
As far as Cat6 and fiber go, generally installation and material costs aren't worth it. We justified it in our composite cable because it's cost wasn't as much. We didn't terminate it yet because so far there's nothing that will use it.
Yup - they have an excellent track record for releasing their source:
The LGPL says the changes have to me made available, possibly at a charge for media. CodeWeavers provides it on the CrossOver CD if you buy it. Or, you can download it.
Then, they have most of their changes integrated into the main WineHQ tree - something they don't have to do. Some of their hackier patches don't necessarily make it in (though you can still get them).
Their tools are not bound by LGPL, but they have released some of them.
At one time SCO dominated the low-end Unix market. But it seems like sometime around 1997 development ground to a halt. What happened? Were all bets hedged on Unixware? OpenServer was pretty much useless without the Skunkware add-on CD. Why is SCO now whining about a product they themselves seemed to have given up on years ago?
Although I wasn't personally there, some friends of mine have a great story about an NEC NEAX PBX. It's probably about the size of 5 refrigerators. We're talking a huge system, supporting thousands of phones, tons of trunks, and a sh*tload of power supplies.
Anyway, the basement of a hospital flooded with this thing in it. The PBX was under several feet of water for a while. Oh, did I mention they hadn't done a backup in months?
After removing the water, they carefully dried everything out. The reconnected power and were able to fire it up (probably just the CPU module) just long enough to get a full backup. From there it was "simply" a matter of installing a new switch and restoring from backup.
Wine can't do anything the normal user couldn't do. Now, a user might not want to mailbomb all of his friends,etc, but the virus still runs with the user's rights. It's not going to delete his whole hard drive (unless he has the rights to) and it's not going to infect system binaries. So At least there's a little bit of protection compared to normal Windows.
I do telecom too, and from what I've seen you've got a pretty unique setup. From what you've described you have some completely custom software. Any chance of unleashing it on the world?
Modern PBX's still run Unix, not that you'd ever know it. But a PBX is only part of the telephony process. But surrounding a PBX are tons of computers - most of which don't seem to run Unix. Voicemail systems, automatic call distribution, billing, etc, etc, seem to use Windows. Even the MAT terminals used to access the PBX are Windows or DOS based.
I can think of several reasons why this won't be classed as a planet:
The chance of this thing having an atmosphere is doubtful. Even Pluto's freezes for most of it's orbit. (Though for the next 15 years it will likely have an atmosphere of nitrogen, methane, and carbon monozide.. then it'll freeze solid again. That's why we should be sending a probe there now.)
It's only 25% bigger than Varuna - another Kuiper Belt object, and probably about the same size as other ones. The unique thing is this one reflects more light and makes it easier to study.
The Kuiper Belt, although interesting, doesn't seem to hold any objects that are substantially different from one another. This is just another one, albeit a hell of a lot bigger.
Before people go trashing on WordPerfect, let me point out some things you might not know about it:
They've supported Unix platforms for a hell of a long time. SCO, HP-UX, Solaris, etc.
They still sell a character-only interface for people who use terminals - useful in large organizations; useful for people who want to maintain compatibility with older versions.
The legal profession still relies on it - your lawyer uses WordPerfect and most legal forms are available in that format. And we all know, once a lawyer makes a document it never goes away.
At times they've had one of the best commercial apps for Unix - print spool manipulation, import/export, spellcheck, desktop publishing, etc. (Although, from release to release some things became dated.)
And if you say it's not for you, you're right. It definitely fills an important niche that a lot of other apps can't or don't want to.
I can think of four things, partly already mentioned:
Document the hell out of it.
As part of that, give clear examples of everything. If possible, build a demo available online.
Provide hooks.. this is somewhat sad, but there have been an occasion or two where I've chosen a product because it had a perl interface or commandline toolset. If I can hack a quick test I'll know within a few hours if I'm going to use the product.
Provide three sets of marketing materials:
"Look how shiny this is." (marketdroid speak)
"Look how shiny this is, and it washes your windows too." (IS Manager speak)
"Look how shiny this is. It washes your windows because it's made of concentrated form of nitric acid." (Developer speak)
I'd almost argue that it's useful to provide the developers with a developer community and mailing list too. But that can backfire if you set it up and choose to ignore it.
I've often wondered how much of it I've lost. I had algebra beaten into my head. Then I got a BS in electrical engineering and had calculus, linear algebra, etc beaten in my head. But recently I've gone through old college text (from 6 years ago) and realized I've lost most of the calculus I knew. At UMich calc was divided into 4 basic classes (calc 1, calc 2, vector calc, and diffy eq). I know I've lost the last two for sure. I could probably do some basic calc (at least I understand the concept still.) But I wonder if I've lost the "basics".
Anyone else wondered this? Have you tested yourself? I took the math section of the GMAT and rocked it, but that's more of a problem solving test.
IANAL: I am not a lawnmower. I'm also not a lawyer.
Given that someone out there cares more than I do (I did care for the first two years, now I want a decision to stick), could someone please explain what will happen if the judge rules in favor of the states? Can Microsoft appeal? What if Microsoft wins? Can the states appeal?
I've recently been going through the building process. I have a GC, but it's definitely a special case so can't recommend for or against having one. I can however tell you how to cut some corners to save yourself some big cash:
Having friends who specialize in different areas is really important. For instance, if you know someone who does drywall, have them subcontract that work and be their labor. Likewise, if you know a good trim carpenter beer and pizza will go a long way. Plus you'll get better materials and work.
Even if you don't personally know the subcontractor, offer him some cash on the side for "extras". Some of those little extras you'd like such as more electrical outlets or an extra gas line on the deck for a grill cost contractors little more than time. They might be willing to work a few extra hours on the side for it. In my case, I got half of my basement drywalled for free, the other two walls are so easy I'll do it myself after the CO.
Beer and pizza. Has anyone mentioned that? Go onsite at least once a week even if you have a GC. Especially when important things like trim work, plumbing, and carpeting are being done. Beer may not be appropriate, but taking lunch orders always is.
Wait until after the CO for anything you don't need right away. Chances are a building inspector will be a lot more lenient if there's not much to inspect. For instance, I want a storage space under the stairs. Temporarily it's being walled in, but later I'll cut some studs and hang a door. Also, I'm saving money by not putting a deck in right now. I have approved plans for one, after the CO I have a friend lined up to put it in - $300 + materials, that's $1000 less than what I'd pay right now.
Don't get bogged down in the details if it starts costing you money. I made some changes to the plans that cost me a few extra bucks, in retrospect they were worth it but had I done all the stuff I wanted I'd be hating life right now. The little extras cost a lot in the long run.
Extra credit is awarded if you know the building inspector. I don't but I can think of a few instances where it would have been invaluable.
I remember working on an RFP that involved some DSN communication. From what I remember, the 70m dishes are going away. Yeah, a huge dish kicks ass, and the amount of bandwidth you can get is phenomenal (hundreds of kilobits on an interplanetary mission!!). However, you can get pretty much the same bandwidth by using the smaller 34m dishes in parallel. The 70m dishes were the old NASA approach of using a sledgehammer on a thumb tack. Modern electronics lets you do some fancy stuff, and it's a lot easier to install a few racks of equipment and use smaller dishes then keep the mechanics on the 70m dishes running. We're likely to not see many infrastructure improvements on the current DSN soon - they did a huge upgrade in the early/mid nineties.
Another real problem is acquiring enough bandwidth at the right time of day. DSN has three main locations because this little blue sphere we're on spins so fast. Scheduling all that communication isn't exactly easy.
Where was Slashdot when the huge license debate was going on a few months ago? It was far more interesting and from what I remember it got one or two mentions. Also Bruce Perens said something about licensing and possibly "Wine Magazine" getting it wrong - I urge him to go back to issues 111, 115, and 116 of Wine Weekly News - I'm pretty sure I got it right. (If I didn't let me know and I'll include the necessary changes.)
I'd also like to point out that Hidenori Takeshima never cited the DMCA as the reason the code was pulled, although ostensibly it's the only legal reason. For all we know his employer could own some of the code.
Anyway, Gavriel State's proposal is pretty interesting because there hasn't been a major sync with the main (LGPL) Wine CVS in a while. Both sides have a lot to offer. The DLL separation is very important to Transgaming - without it their work will become horribly out of sync. Likewise, the DIB engine and DCOM code would be nice stuff to have in the main Wine tree. (Although, if you read closely it says "current work in WineX that supports DCOM" - according to Ove Kaaven he's work on some new and improved stuff that will make the "current work" obsolete. Perhaps his new stuff has already hit the WineX tree and I'm wrong, but my hunch says they haven't finished it yet.)
Has anyone ever heard of a BSD and LGPL project trading patches back and forth? I'm sure example exist, perhaps some of the Linux and FreeBSD drivers have worked that way in the past.
I spent the majority of my day yesterday babysitting a T-BERT. We're having a mysterious problem with a voice T1 that has channels drop off occasionally. All the vendors are pointing fingers at each other.
Anyone happen to know any other good resources for troubleshooting T1's? The Cisco link above looks interesting.
Slashdot Mirror
I went through almost the same situation.. granted it was 10 years ago and at the time no one heard of something like a network engineer..
First off, I think a college degree can open a lot of doors. I'm fortunate enough to be doing exactly what I love in a part of the country most people can't afford to live in year round. I suspect I couldn't have done that with my resume alone. I can guarantee you won't get the hands on experience in an undergraduate program that someone getting a CCNA will get, but I don't think you should worry about that.
Now, getting down to some details. I think an Electrical Engineering, Computer Science, or Computer Engineering degree is what you should be looking at. I realized about my junior year I wanted to get into networking stuff and found I had some great resources to choose from:
My first job was a sys admin. Excellent stuff and it gave me firm hands-on training. Now I do telecommunications, which mainly involves phone networks and infrastructure backbones for the computer geeks (fiber and copper in a campus environment.) I firmly believe you don't know shit about computer networks until you've worked on phone networks - it's kind of the missing piece that makes everything else make sense.
One other random thought.. someone told me once that college isn't about teaching you a profession it's about teaching you how to learn.
Good luck!
Actually, I haven't been a Linux sys admin for almost 4 years now. I did a nice 4 year stint admining Linux/Unix back when getting a job working with Linux was next to impossible and Knoppix didn't exist.
I still do that work on the side and still use Linux for personal use. But rather than having to admin 50 - 60 Linux servers I'm down to about 2. With a little luck I'll never have to do a rescue again. I've heard good things about Knoppix. Based on what you said maybe I should download it just to boot it up. Thanks!
I can't say I really care one way or the other about this.
Novell has been quite stagnant for a while. Thankfully they realized that they needed a shot in the arm in order to expand their increasingly legacy user base. But have they done anything interesting? Will they be a contributing member of the community? Why should I get excited about Novell?
Then you have SuSE. I've used the distribution quite a bit. More often than not for their Live CD to do rescues. It's an ok distribution - it definitely has more software than any other distro and the book they provide is excellent. But again, it's not that interesting of a distro. It's just enough of a pain in the ass to use to not be enjoyable.. unless you think Slackware combined with RPM's is cool or you enjoy finding random documentation written in German.
Can someone please tell me something I don't know that will make me care about two also-rans?
Um, I didn't really write anything about the reactions of Wine developers in this weeks' WWN. Save for agreeing with Mike Hearn that an announcement still seems pending.
Oh well, thanks for the press. It's always nice to see.
A lot of people have already mentioned the PhD-is-good-for-research theory, so I won't rant on that.
I've seen a lot of people lately who are staying in school rather than throwing themselves at a tough labor market. Personally, I think staying in school to escape "the real world" is a really weak excuse. If you've only been in school and don't have practical experience then I highly suggest getting out and getting a job. I don't get along as well with my friends who are still in school (I graduated five years ago with a BS). A lot of them seem to be in perpetual procrastination about putting together a resume, interviewing for jobs, and in constant pursuit of the next easy part-time job.
Now, having said that, perhaps you have some personal goals you'd like to accomplish or a certain area of study you really want to pursue. Doing that in a non-academic environment can be rough and then it makes sense to stay in school.
There's lots of exciting things to do in this world besides working on a thesis. While I think pursuing a career (you do have a grad degree) is good, I'm sure you could come up with something more original. Go work on a cruise ship for a year. Take off and lay on the beach in Thailand for a while. Spend the winter skiing.. of course that's what I did and accidentally never left.
If you're a tranport ISP and not on the edge I just don't see what options there are (were). Known exploits were in the wild and the only solutions are either Cisco's patch or something ACL based. ACL based might be fine if you can tolerate the increased load. Otherwise, the patch is it. And that sucks if you have to do something like cross the 12.0(22)S hurdle.
Of course you should be careful, but if you're working in a large NOC and making the big bucks this is the type of thing you stay at work for hours testing. Yes, this exploit will wreck havoc for years, but hopefully not on any backbone provider's network. It's the time to stock up on caffeine and tell your wife you'll be home late. With all the media attention most networks put out customer notifications of impending outage and I think most people thought it was ok.
As far as the Linux references (especially from the other guy), I don't buy it either. Until someone makes a nice canned Linux router product with Cisco style support it's simply not a solution. As far as Juniper goes, I think this is an excellent example of why it's a good thing to have mutiple router platforms in a datacenter. Of course I've said the exact opposite when it comes to having to support it..
Please don't assume a NOC is a large network scaled up. There are a million reasons why that's not the case. (sorry, couldn't resist)
There's no choice in the matter. This vulnerability has one fix available from Cisco (with the load raising ACL solution a distant 2nd) with known exploits in the wild. Very little extensive testing of this patch has been done and no one knows if the patch will break anything else. (Supposedly Cisco does 250,000 internal tests before releasing a patch.) Yes, there's unconfirmed reports of inout errors resulting from the patch.
If you are a transit ISP you test this as much as possible and roll it out. It's a game of Russian roulette and you're forced to play. If you're sitting in a mission critical NOC you apply this, deal with the fallout, and go to plan B in the event of something bad. That's why you're getting paid the big bucks, right?
If you're on the edge you probably have support to fall back on and in this instance it's exactly what you'd want to do.
Cisco released the fix two days ago to backbone providers. Other large customers could get the fix early yesterday. If you're affected by this vulnerability and it's not fixed yet:
It seems like Cisco handled this one correctly with the providers. I'm not sure how well large customers were handled, my guess is the .edu folks probably got screwed again.
I think it'd be quite important to get the FCC involved to regulate the frequencies.
We've been putting in some new hotel/condo buildings lately. Specifically we've built 8 new ones comprised of 500+ units. We made the decision to pull composite cable everywhere. We chose 2 RG-5, 2 Cat 5, and 2 strands of fiber in each cable. We used Siemen's "Home Cabling System" for all the terminations. Good stuff.
As for your existing building, recabling will be expensive. If you're lucky there's riser closets on each floor where you could use the existing cable to pull with. You could also locate hubs there. But that's a lot of work. The idea of using some low bandwidth solution to the unit isn't too bad. Hardware costs will be really high, but you may be able to pass it along to the homeowner rather than eat it yourself. Yeah, it'd be nice to have ethernet everywhere but that just might not be possible.
As far as Cat6 and fiber go, generally installation and material costs aren't worth it. We justified it in our composite cable because it's cost wasn't as much. We didn't terminate it yet because so far there's nothing that will use it.
At one time SCO dominated the low-end Unix market. But it seems like sometime around 1997 development ground to a halt. What happened? Were all bets hedged on Unixware? OpenServer was pretty much useless without the Skunkware add-on CD. Why is SCO now whining about a product they themselves seemed to have given up on years ago?
Although I wasn't personally there, some friends of mine have a great story about an NEC NEAX PBX. It's probably about the size of 5 refrigerators. We're talking a huge system, supporting thousands of phones, tons of trunks, and a sh*tload of power supplies.
Anyway, the basement of a hospital flooded with this thing in it. The PBX was under several feet of water for a while. Oh, did I mention they hadn't done a backup in months?
After removing the water, they carefully dried everything out. The reconnected power and were
able to fire it up (probably just the CPU module) just long enough to get a full backup. From there it was "simply" a matter of installing a new switch and restoring from backup.
Wine can't do anything the normal user couldn't
do. Now, a user might not want to mailbomb all of his friends,etc, but the virus still runs with the user's rights. It's not going to delete his whole hard drive (unless he has the rights to) and it's not going to infect system binaries. So At least there's a little bit of protection compared to normal Windows.
I do telecom too, and from what I've seen
you've got a pretty unique setup. From what
you've described you have some completely
custom software. Any chance of unleashing it
on the world?
Modern PBX's still run Unix, not that you'd ever know it. But a PBX is only
part of the telephony process. But surrounding a PBX are tons of computers - most of which don't seem to run Unix. Voicemail systems, automatic call distribution, billing, etc, etc, seem to use Windows. Even the MAT terminals used to access the PBX are Windows or DOS based.
I can think of several reasons why this won't
be classed as a planet:
for most of it's orbit. (Though for the next 15 years it will likely have an atmosphere of nitrogen, methane, and carbon monozide.. then it'll freeze solid again. That's why we should be sending a probe there now.)
reflects more light and makes it easier to study.
Before people go trashing on WordPerfect, let me point out some things you might not know about it:
And if you say it's not for you, you're right. It definitely fills an important niche that a lot of other apps can't or don't want to.
I can think of four things, partly already mentioned:
I'd almost argue that it's useful to provide the developers with a developer community and mailing list too. But that can backfire if you set it up and choose to ignore it.
I've often wondered how much of it I've lost. I had algebra beaten into my head. Then I got a BS in electrical engineering and had calculus, linear algebra, etc beaten in my head. But recently I've gone through old college text (from 6 years ago) and realized I've lost most of the calculus I knew. At UMich calc was divided into 4 basic classes (calc 1, calc 2, vector calc, and diffy eq). I know I've lost the last two for sure. I could probably do some basic calc (at least I understand the concept still.) But I wonder if I've lost the "basics".
Anyone else wondered this? Have you tested yourself? I took the math section of the GMAT and rocked it, but that's more of a problem solving test.
IANAL: I am not a lawnmower. I'm also not a lawyer.
Given that someone out there cares more than I do (I did care for the first two years, now I want a decision to stick), could someone please explain what will happen if the judge rules in favor of the states? Can Microsoft appeal? What if Microsoft wins? Can the states appeal?
I've recently been going through the building process. I have a GC, but it's definitely a special case so can't recommend for or against having one. I can however tell you how to cut some corners to save yourself some big cash:
Extra credit is awarded if you know the building inspector. I don't but I can think of a few instances where it would have been invaluable.
I remember working on an RFP that involved some DSN communication. From what I remember, the 70m dishes are going away. Yeah, a huge dish kicks ass, and the amount of bandwidth you can get is phenomenal (hundreds of kilobits on an interplanetary mission!!). However, you can get pretty much the same bandwidth by using the smaller 34m dishes in parallel. The 70m dishes were the old NASA approach of using a sledgehammer on a thumb tack. Modern electronics lets you do some fancy stuff, and it's a lot easier to install a few racks of equipment and use smaller dishes then keep the mechanics on the 70m dishes running. We're likely to not see many infrastructure improvements on the current DSN soon - they did a huge upgrade in the early/mid nineties.
Another real problem is acquiring enough bandwidth at the right time of day. DSN has three main locations because this little blue sphere we're on spins so fast. Scheduling all that communication isn't exactly easy.
Where was Slashdot when the huge license debate was going on a few months ago? It was far more interesting and from what I remember it got one or two mentions. Also Bruce Perens said something about licensing and possibly "Wine Magazine" getting it wrong - I urge him to go back to issues 111, 115, and 116 of Wine Weekly News - I'm pretty sure I got it right. (If I didn't let me know and I'll include the necessary changes.)
I'd also like to point out that Hidenori Takeshima never cited the DMCA as the reason the code was pulled, although ostensibly it's the only legal reason. For all we know his employer could own some of the code.
Anyway, Gavriel State's proposal is pretty interesting because there hasn't been a major sync with the main (LGPL) Wine CVS in a while. Both sides have a lot to offer. The DLL separation is very important to Transgaming - without it their work will become horribly out of sync. Likewise, the DIB engine and DCOM code would be nice stuff to have in the main Wine tree. (Although, if you read closely it says "current work in WineX that supports DCOM" - according to Ove Kaaven he's work on some new and improved stuff that will make the "current work" obsolete. Perhaps his new stuff has already hit the WineX tree and I'm wrong, but my hunch says they haven't finished it yet.)
Has anyone ever heard of a BSD and LGPL project
trading patches back and forth? I'm sure example exist, perhaps some of the Linux and FreeBSD drivers have worked that way in the past.
I spent the majority of my day yesterday babysitting a T-BERT. We're having a mysterious problem with a voice T1 that has channels drop off occasionally. All the vendors are pointing fingers at each other.
Anyone happen to know any other good resources for troubleshooting T1's? The Cisco link above looks interesting.