Um, carbon fiber is used quite a bit in racing sailboats. Carbon fiber masts and spinnaker poles are very common these days. There are growing numbers of racing boats with carbon fiber hulls. I think this is a well understood issue, and carbon is quite safe in salt water. Most boat owners use a bottom paint, whose purpose is primarily to retard growth on the bottom. More and more boats are going naked on the bottom though, in lieu of a toxic bottom paint. Me, I still use VC-17 with a bag of pure copper dust mixed in.
Perhaps you should re-evaluate what you question. The US Constitution doesn't grant or create any rights. That was true before there was a Bill of Rights, and is no less true afterward. It merely recognizes them. That's a great distinction. We believe that people have certain inalienable rights. And our Constitution recognizes those. And per the 9th Amendment, its specific recognition of a very small subset of our rights does not imply that we do not have more. Notwithstanding that the Supremes historically don't like the 9th Amendment and would prefer to find asinine things like 'penumbras' of other rights.
We the people are sovereign, we hold all power, and we have all rights. My rights don't come from a piece of paper, a court, or Congress, or my neighbors.
Regarding the right to keep and bear arms: there are those, such as I, who would argue that a free person has that right, regardless of the existence of the 2nd amendment. An unfree person does not have that right. A free person has a right to the means necessary to protect his or or liberty, life and property from all enemies, foreign or domestic. The question is not whether we have the right. The question is to what extent can that right be regulated, and that is a good question. And now, the Supreme Court has finally set us on the path of answering THAT question, not debating over whether we have a fundamental right or not.
To the point of not needing guns: we need arms to protect ourselves from a tyrannical government. I'm not saying that we need to overthrow our government now or at any foreseeable time, or even that we could. I am saying that as free people we have the right to the means to do so, even if the need seems implausibly remote, and a good way to continue to ensure that implausibility is to continue to let free people arm themselves. A people stripped of their fundamental right to protect their liberty, by force of arms if necessary, can only be stripped of more rights. The fact that we retain the right to arms, that we remain vigilant and cognizant of our fundamental rights as free people, is a strong indicator that we retain our other equally important rights.
What on earth are you blabbering about? Yes, Apple has changed, some may even say improved, things: launchd, network configuration, so on. Mounting partitions in the tree is a PIA. There is lookupd. So on. But the OS is *nix, despite all the mucking around. Over the years I've used and developed for SunOS, Ultrix, DYNIX, UNICOS (vector and MPP), SysV, AIX, IRIX, Solaris, NEXTSTEP, Linux, *BSDs, and several more I've forgotten about (what did the Goulds run?). They are/were all *nix, despite the differences. So is OS X. And I hear that Windows is moving more and more toward the Unix Way, if not becoming yet another *nix. What was it Henry Spencer said?
We support dozens of corporations/entities with in-house Zimbra servers, and host shared Zimbra, for customers who consider the groupware aspect of Zimbra critical. What you say is true for educational installs: many of those do indeed disable the calendaring entirely, even disabling IMAP. But that is certainly not the majority of business in our experience.
As for disaster recovery, Zimbra, with a commercial license, has excellent hot-backup and recovery. So long as you keep your backups elsewhere, you can recover pretty painlessly, in the scheme of things. Zimbra also maintains up to the second transaction logs. If you sync those offsite along with backups, you could in theory lose nothing whatsoever (depending on how often you sync of course) if a server catches fire. If you use SAN based replication to an offsite storage server, you could recover to, or fail over to, a second install. Zimbra is also working on native replication, which will not require storage based replication.
Zimbra also is HA. In our shared environment our storage is on a fibre attached SAN with blade servers attached. We can fail over to a standby blade if need be. Zimbra is entirely self-contained, meaning that the server needs very little locally. You can dismount Zimbra from the failing server (or in reality power it down), mount it on the failover, start Zimbra up, and you're good to go. LDAP, which Zimbra depends on very heavily, is replicated.
Migration from Zimbra: Zimbra will export your mail data in a zip of standard email messages. It will export your calendar in ical. Contacts via csv. Definitely not a single file, but it can be gotten out and work with standard tools. Most data can be extracted via the REST interface.
This all sounds like an advertisement... to be sure, Zimbra has its issues. It seems that they don't do regression testing as well as they could -- what are very obvious issues if only they were tested get into release versions. We run into performance issues at times that need to be tracked down. But the caveat is that Zimbra is constantly adding features and is quite responsive to problems. Zimbra became quite mature in the 4.5.x line. The new 5.x line adds more features and should scale better (just added full service proxying), but also has brought in a lot of new bugs and issues. But they are being addressed rapidly, this new release is maturing quickly.
Ok, first, I'm working heavily with Zimbra these days. We provide hosted Zimbra for thousands of users in a multi-node install, and manage dedicated servers for dozens of companies.
Zimbra absolutely does not use Cyrus IMAP! Zimbra implements web, IMAP, POP, LMTP delivery and more in Java. The current version (5.x) uses Jetty for the servlet container. The new version has native tasks as well.
Dedicated machines: absolutely. Every customer we have provides a dedicated machine, but that can also be a virtual machine. Though Zimbra does not support Xen due to mutex locking issues in the Xen kernel. The "packaging" is a misnomer. Zimbra doesn't package in the sense that you manage it. Those RPMs aren't there to let you upgrade them. Zimbra has its own installer, and installs everything in the/opt/zimbra tree by default. We have tweaked our installs to change how services bind, but not in order to install more non-Zimbra services on production machines. See:
Zimbra is a pig on resources. It demands lots of RAM and very fast disks. However, it also scales very well. Once you meet the need for dedicated hardware with ample resources, you can get a lot of mileage out of it. You can scale out horizontally into a multi-node architecture very easily.
I would not recommend trying to run Zimbra on a machine being used for other tasks, unless you virtualize it (vmware, or a Xen HVM). Not unless you want to delve into the guts of Zimbra, which isn't all that hard really. You can force Jetty to bind to a particular IP pretty easily, so you could at least run httpd on another IP. No problem with Samba. If you wanted to use LDAP, I'd recommend binding another instance to a different IP.
Yeah. I don't agree with the summary. I lived through the 80s, and I remember people, including some small businesses I worked for, ALREADY seeking alternatives to Microsoft. Many tried OS/2 and liked it quite a bit, as did I. Too bad it didn't survive. Here's a funny quote I found: "In my opinion, Microsoft was intentionally making OS/2 as difficult to use as possible - or the programmers they had assigned to write OS/2 were the stupid ones." The "problem" was Microsoft was smart and made developing for DOS and Windows very easy, not that the IBM imprimatur hurt to start out. The momentum of application development got going on Windows, which was "good enough," and hasn't ever stopped.
Yes. We saw this today. Turns out Postini still used ORDB as of earlier today. This was reported (hostname changed a bit for privacy):
----- Transcript of session follows -----... while talking to postinicustomer.foo.s7a1.psmtp.com.: >>> RCPT To:<user@postinicustomer.foo> <<< 550 64.18.2.63 blacklisted at relays.ordb.org
One would think that they OF ALL PEOPLE would know better!
Under the highly regulated AT&T regime that provided legendary reliablity, equipment was not cheap, and capacity was not plentiful. I'll wager you are way too young to remember. AT&T used to provide the phones. It was illegal to attach your own to the network. You couldn't go to the store and buy one. You couldn't get cordless phones. Only once that rule was changed did we start to see innovation in phones -- AT&T had no interest in giving you more than Princess. Likewise telecommunications. When I was young I had a 300 baud modem. Long distance rates were expensive... the BBSes/hosts of the day used a cell-network like method (recall Fido? UUCP? bang addressing?) to get email across the country. It took hours or more. Cell phones -- originally highly regulated and expensive (my Dad even had a wireless phone, precursor to the cellular network), with a single competitor to the incumbent telco permitted per market.
How did we get cheap cell phones, and a plethora of variety and capabilities? By competition, starting with the PCS spectrum auctions that ended the A-B carrier regime. How did we get the Internet? We got it when the federal government contracted with private companies to provide the backbone for the NSFnet, formerly the ARPAnet, and then began permitting interconnections. We got it through competition and lack of strict regulation. How did network equipment get so cheap and plentiful, so much fibre laid? You guessed it. Do you really think that if we had not had the environment created by that, the goldrush of the laissez faire Internet in the 90s, that networking equipment would be so cheap, that there would be such innovation, or that those "other countries" would have been able to ride on the coat-tails of that?
Fast, cheap, good. I've worked in the super-computing industry, for Motorola and some other telecom companies after that, and for several networking/Internet based companies before, during, and now after the boom. Trust me, it has always applied to far more than just manufacturing widgets.
"Comcast and the rest can provide almost no service at all for what we pay them and they get away with it."
Note, "what we pay them." We pay them prices based on competitive forces, where reliability is just one factor. Granted, Comcast may not be the best example. But think in general.
The way the phone network got so reliable was because we granted a monopoly, and granted guaranteed, predictable profits. If it cost X to get the standards required, fine -- it was paid for, and there were *always* profits on top. That is key. We cannot have our cake and eat it too. Fast, good, cheap. We all know it. Look at the power companies. We have politicians pushing populist agendas to freeze rates, limit profits -- and the result is that the private companies running the power grid simply cut back elsewhere, and we have power outages and very little new power generation (of course that is also to be blamed on NIMBY opponents).
Perhaps we should have some more oversight of the Internet today, some sort of oversight board. But even if we did have that, would it or could it prevent the peering spats? Should it? What would a review board do in the case of an American company and European company with a contract that wasn't being honored? Would we need some sort of United Nations entity? How would this impact innovation and interest if decisions had to be brought before regulatory entities, subject to public comment, so on. If companies simply cannot depeer and make it actually hurt (what Cogent did by blocking traffic), then where do the incentives come from to provide the peering agreed upon? If we regulate it and mandate reliability, will we also regulate and mandate paying the true cost, along with a healthy enough profit to make sure a private entity remains interested in maintaining the network and providing for future growth and capacity? What will lack of competition do to cost, and market penetration? Will regulation drive away the private investors who fund these companies in hope of turning a healthy profit? Will we all pay for it via higher service fees or taxes? Is it perhaps ok to have these occasional spats, if the end result is a reasonably robust network at the "best" price? Or should the whole thing just be one big government funded and controlled system paid for by taxes and usage fees? Do we trust the government? Where would the innovation come from?
Perhaps what we have is the best of many imperfect possibilities?
Um, I used to use this Internet thingy when it was ARPAnet, before the advent of private backbones. I remember HOSTS.TXT and the real InterNIC. And yes, it was originally designed to route around major failures. That was one of the reasons DARPA, e.g. the military, funded it. It may have not done it perfectly, it may not have been able to survive a full-on nuclear conflict, but it was certainly designed and funded in good part as a research project into network robustness in the face of catastrophe.
Ever since the backbones went private though, all bets are off. You are entirely correct as of the early 90s. As we all know, it's "my network, my rules." Hence this peering spat, and the ones before, and the ones to come.
> I love it how Randians try to blame corporate corruption on governments.
I don't know whether I'm a "Randian" but I know that corporations are artificial legal entities that exist only by the consent of the government, legal entities that are essential. Those who rail against the existence of corporations, using the word "corporate" as an epithet, are no better than those on the other end of the spectrum. Corporations are artificial entities -- they cannot be evil, only people can be evil.
Government is certainly partially culpable for the worst corporate excess when the environment for that excess was greased into existence by money finding its way into the hands of corrupt politicians of all political bents.
In your extreme example of a monopoly on food production and distribution, the only way that could happen is if government did not work properly in its oversight of that legal construct. It isn't the corporation that is evil, it is the people involved in the little drama who are evil. The people who run the corporation, the people who work for it doing evil things, the shareholders who learn of the evilness and do nothing, the police paid to not intervene, inspectors, judges, politicians, so forth. I guarantee that all of those would be involved in the creation of such a scenario. They are the evil ones. And ultimately the people themselves, because what else is "government" than a collection of people, for permitting that environment to exist. To simply rail against "evil corporations" is childish. The road was well traveled.
So I don't think the original premise is entirely correct -- the biggest, nastiest corporation can *force* you, but only after it *enticed* a lot of people, including those in government, to permit it to become big and nasty. So where does the blame fall? And who is evil?
After this debacle, Democrats need to figure out what their goal is. While their system may be "fair" it is also fairly stupid. The party needs to get a candidate and not excessively damage that candidate in the process of choosing him/her/it. They should either have all states vote in a very shortened time span, or move to a winner take all system. Instead, this thing will drag on due to their short-sighted but "fair" proportional allocation of delegates. That may be fine in a real election, done on one day. But not like this, dragged out over months. It's idiocy.
Not to mention: why is the taxpayer footing the bill for any portion of the primary process? These political parties are private clubs. How they choose their candidates should be their business, using their rules, and on their dime. Not only should the law not mandate "open" primaries, the law should not mandate caucus versus secret ballots, nor treat these any differently than your local Bingo club electing officials. The election is in November.
Your anecdotal experience is one thing, but facts are facts. Having said that, this is much like a LP vs CD vs MP3 discussion.
I worked for Motorola in the Cellular Infrastructure Group back in the mid 90s, the period when PCS license blocks were being auctioned off by the FCC and we began the transition from a dual-carrier marketplace to our modern multi-carrier marketplace in the US -- I later went to work at one of those PCS license-holders for a bit. I'm a bit dated on some things in the industry, but the fundamentals at the transmission level haven't changed all that much. AMPS exclusively dedicates a fixed amount of bandwidth from frequency A to B to your voice, an amount derived from the channelization of POTS lines. That was FDMA -- the total allocated bandwidth was split into two full dedicated audio channels per conversation, with no overlap, chosen primarily to match POTS voice bandwidth and quality. N-AMPS came later and sacrificed some bandwidth, but still provided plenty for the voice. TDMA is digital sampling and packetization of the voice, sorta like VOIP these days. The primary design criteria for TDMA was to increase bandwidth utilization with an acceptable loss of voice quality, along with providing additional features. When I was at Motorola there were guys working in a lab nearby with Qualcomm on the early CDMA technology -- it was in theory capable of providing voice quality on par with AMPS, but like TDMA one of its primary goals is utilization with acceptable quality degradation. It didn't work so great in those early days, but is now a superior technology for voice quality I believe. In fact next generation GSM is an amalgam with CDMA mixed in.
That's just the raw technology though. Implementation is key. All things being equal, we are not likely to get better pure voice quality on a consumer cellular network than was had back in the days of AMPS with analog engineers pursuing voice quality as job one, not capacity. There are too many cell phones now for that to be the goal.
That TDMA thing always irks me. Current GSM is TDMA, you know (though next generation will be CDMA). It means Time Division Multiple Access, essentially packetizing data. So while some forms of TDMA may be obsolete, the world's most popular mobile phone system also is a form of TDMA.
I'm sorry, but better sound? AMPS with a clear signal is the best you are going to get. AMPS, a form of FDMA, was designed back in the days of just an A carrier and B carrier per market, when cell service was expensive. AMPS, and even N-AMPS, wasn't designed to maximize calls per antenna, it was designed to match the voice bandwidth used by analog telephone lines, essentially a wireless equivalent. Everything since then has been designed to maximize capacity at the expense of voice quality. PCS, TDMA, GSM (which is TDMA BTW -- TDMA is not analog), CDMA, all have as a major design criteria increasing the call capacity of limited frequency bandwidth. The result is greater competition, reduced cost, wider availability of cell phones, everybody and their pet carrying one, etc., all good things perhaps -- but "better" sound quality? I don't think so.
Good point. But it's part of what I consider the order of things to do. My obligation as a buyer is to pay for the item promptly or as otherwise agreed. My obligation as a seller is to sell the item I described and ship it as agreed. Once the buyer fulfills their side, the seller should say so. Once the seller fulfills their side, the buyer should say so. Somebody has to go first, unless your idea of blind feedback, which I do like quite a bit, were implemented. I feel it is unreasonable that the buyer leaves feedback first, but is fearful of leaving honest negative feedback for fear of an unwarranted retaliatory negative response. Sellers tend to have more leverage in the system than buyers, particularly in this day and age where so many sellers are businesses/quasi-businesses.
And yes, whatever. I've always done it this way, it's my policy (even published on that About Me page). Sometimes as a buyer I never get feedback, because the seller never leaves it, even when I do tell them that all is well and I'm happy. I've been told by many that they have an automated feedback system that leaves positive feedback only once I do. Ridiculous. And yes I run some risk of getting a negative feedback after I, as seller, already left mine once the buyer paid as agreed -- but that's the risk you run. I'm honest, I charge fairly for shipping, and I have 100% positive feedback over 10 years, and have never had a bad ebay experience. Well, I take that back. Two not so perfect experiences, both involving bubblewrap and insufficient peanuts. In both cases the sellers offered to make good for the damage.
I disagree. As a seller, I leave feedback once the person pays, as a rule. That is their obligation to me. They pay, I ship. If they complain via feedback, I can respond. But I never, ever hold feedback hostage. And likewise, as a buyer, I never, ever leave feedback until the seller does.
And I certainly have noticed in recent years how many sellers do not leave feedback until they see what you have to say. Unfortunate. But I'm iron-clad in my policy on feedback, I've been doing it that way for ten years now.
Also, anybody remember when you could put comments on an active auction, e.g. "are you crazy, you buy that widget for $X brand new in the store!" I hope they bring that back. Ebay has gotten way too seller/merchant biased.
No, there are lots of cars with transverse engines that are home-mechanic friendly. My Dad's a Buy American kind of guy with cars, and his cars have always been more difficult to work on that my Japanese cars, mainly Nissan and Mazda. Seems most American cars expect you to be using a mechanic with a lift for everything. I remember on one of my Dad's American cars, trying to change his oil one time with a floor jack in the driveway... it wound up being easier to take off a wheel to get access to the filter than do it from below on the ground.
Regarding the add to cart -- no, that is not accurate. I just searched for a list of domains -- just searched -- and all are now held by NSI. However, it holds only the.com version, although I did not provide a TLD in my query list. I am so writing a program to automate this. It appears you have to feed it in through their web CGI, a whois query doesn't result in the hold.
How about the one where the government can now IN SECRET, presenting no evidence or holding no hearing or trial as GUARANTEED by the Constitution, seize your assets? How is that "protecting the constitution?
This does seem a rather natural progression given what we've been doing to our rights for A LONG TIME. Ever heard of civil forfeiture? It's been in use for a long time before Bush.
Gotta love cases like United States v. One Assortment of 89 Firearms.
Bush may be doing things that are quite objectionable to quite many, but the funny thing is, things that would have been clearly unconstitutional in the past (such as, oh, a federal minimum wage, or federal workplace safety laws) suddenly become constitutional when expedient, with no substantive changes in the underlying text or intent. I always tell people, a constitution subject to interpretational whim is a slippery slope. If you let politicians and judges decide how our fundamental legal foundations change without the bother of amendments and the entire associated process, then odds are it will change in ways you both like and do not like. You can't have it only your way. Unless you try to pack the Court. Like FDR.
Fire trucks are fun. My eccentric dad has a '55 Mack pumper, and a '40-something 100' open cab American LaFrance ladder. I've run a red light or two... it's a pain to get that pumper going, straight cut gears and all. We need to get them running again this summer.
Slashdot Mirror
Um, carbon fiber is used quite a bit in racing sailboats. Carbon fiber masts and spinnaker poles are very common these days. There are growing numbers of racing boats with carbon fiber hulls. I think this is a well understood issue, and carbon is quite safe in salt water. Most boat owners use a bottom paint, whose purpose is primarily to retard growth on the bottom. More and more boats are going naked on the bottom though, in lieu of a toxic bottom paint. Me, I still use VC-17 with a bag of pure copper dust mixed in.
Larry
Perhaps you should re-evaluate what you question. The US Constitution doesn't grant or create any rights. That was true before there was a Bill of Rights, and is no less true afterward. It merely recognizes them. That's a great distinction. We believe that people have certain inalienable rights. And our Constitution recognizes those. And per the 9th Amendment, its specific recognition of a very small subset of our rights does not imply that we do not have more. Notwithstanding that the Supremes historically don't like the 9th Amendment and would prefer to find asinine things like 'penumbras' of other rights.
We the people are sovereign, we hold all power, and we have all rights. My rights don't come from a piece of paper, a court, or Congress, or my neighbors.
Regarding the right to keep and bear arms: there are those, such as I, who would argue that a free person has that right, regardless of the existence of the 2nd amendment. An unfree person does not have that right. A free person has a right to the means necessary to protect his or or liberty, life and property from all enemies, foreign or domestic. The question is not whether we have the right. The question is to what extent can that right be regulated, and that is a good question. And now, the Supreme Court has finally set us on the path of answering THAT question, not debating over whether we have a fundamental right or not.
To the point of not needing guns: we need arms to protect ourselves from a tyrannical government. I'm not saying that we need to overthrow our government now or at any foreseeable time, or even that we could. I am saying that as free people we have the right to the means to do so, even if the need seems implausibly remote, and a good way to continue to ensure that implausibility is to continue to let free people arm themselves. A people stripped of their fundamental right to protect their liberty, by force of arms if necessary, can only be stripped of more rights. The fact that we retain the right to arms, that we remain vigilant and cognizant of our fundamental rights as free people, is a strong indicator that we retain our other equally important rights.
Larry
What on earth are you blabbering about? Yes, Apple has changed, some may even say improved, things: launchd, network configuration, so on. Mounting partitions in the tree is a PIA. There is lookupd. So on. But the OS is *nix, despite all the mucking around. Over the years I've used and developed for SunOS, Ultrix, DYNIX, UNICOS (vector and MPP), SysV, AIX, IRIX, Solaris, NEXTSTEP, Linux, *BSDs, and several more I've forgotten about (what did the Goulds run?). They are/were all *nix, despite the differences. So is OS X. And I hear that Windows is moving more and more toward the Unix Way, if not becoming yet another *nix. What was it Henry Spencer said?
Larry
Sys-admins set up AD rules? I thought that was done by MSCE hacks.
Larry
We support dozens of corporations/entities with in-house Zimbra servers, and host shared Zimbra, for customers who consider the groupware aspect of Zimbra critical. What you say is true for educational installs: many of those do indeed disable the calendaring entirely, even disabling IMAP. But that is certainly not the majority of business in our experience.
As for disaster recovery, Zimbra, with a commercial license, has excellent hot-backup and recovery. So long as you keep your backups elsewhere, you can recover pretty painlessly, in the scheme of things. Zimbra also maintains up to the second transaction logs. If you sync those offsite along with backups, you could in theory lose nothing whatsoever (depending on how often you sync of course) if a server catches fire. If you use SAN based replication to an offsite storage server, you could recover to, or fail over to, a second install. Zimbra is also working on native replication, which will not require storage based replication.
Zimbra also is HA. In our shared environment our storage is on a fibre attached SAN with blade servers attached. We can fail over to a standby blade if need be. Zimbra is entirely self-contained, meaning that the server needs very little locally. You can dismount Zimbra from the failing server (or in reality power it down), mount it on the failover, start Zimbra up, and you're good to go. LDAP, which Zimbra depends on very heavily, is replicated.
Migration from Zimbra: Zimbra will export your mail data in a zip of standard email messages. It will export your calendar in ical. Contacts via csv. Definitely not a single file, but it can be gotten out and work with standard tools. Most data can be extracted via the REST interface.
This all sounds like an advertisement... to be sure, Zimbra has its issues. It seems that they don't do regression testing as well as they could -- what are very obvious issues if only they were tested get into release versions. We run into performance issues at times that need to be tracked down. But the caveat is that Zimbra is constantly adding features and is quite responsive to problems. Zimbra became quite mature in the 4.5.x line. The new 5.x line adds more features and should scale better (just added full service proxying), but also has brought in a lot of new bugs and issues. But they are being addressed rapidly, this new release is maturing quickly.
Larry
Ok, first, I'm working heavily with Zimbra these days. We provide hosted Zimbra for thousands of users in a multi-node install, and manage dedicated servers for dozens of companies.
/opt/zimbra tree by default. We have tweaked our installs to change how services bind, but not in order to install more non-Zimbra services on production machines. See:
Zimbra absolutely does not use Cyrus IMAP! Zimbra implements web, IMAP, POP, LMTP delivery and more in Java. The current version (5.x) uses Jetty for the servlet container. The new version has native tasks as well.
Dedicated machines: absolutely. Every customer we have provides a dedicated machine, but that can also be a virtual machine. Though Zimbra does not support Xen due to mutex locking issues in the Xen kernel. The "packaging" is a misnomer. Zimbra doesn't package in the sense that you manage it. Those RPMs aren't there to let you upgrade them. Zimbra has its own installer, and installs everything in the
Zimbra is a pig on resources. It demands lots of RAM and very fast disks. However, it also scales very well. Once you meet the need for dedicated hardware with ample resources, you can get a lot of mileage out of it. You can scale out horizontally into a multi-node architecture very easily.
I would not recommend trying to run Zimbra on a machine being used for other tasks, unless you virtualize it (vmware, or a Xen HVM). Not unless you want to delve into the guts of Zimbra, which isn't all that hard really. You can force Jetty to bind to a particular IP pretty easily, so you could at least run httpd on another IP. No problem with Samba. If you wanted to use LDAP, I'd recommend binding another instance to a different IP.
Larry
Yeah. I don't agree with the summary. I lived through the 80s, and I remember people, including some small businesses I worked for, ALREADY seeking alternatives to Microsoft. Many tried OS/2 and liked it quite a bit, as did I. Too bad it didn't survive. Here's a funny quote I found: "In my opinion, Microsoft was intentionally making OS/2 as difficult to use as possible - or the programmers they had assigned to write OS/2 were the stupid ones." The "problem" was Microsoft was smart and made developing for DOS and Windows very easy, not that the IBM imprimatur hurt to start out. The momentum of application development got going on Windows, which was "good enough," and hasn't ever stopped.
Larry
Yes. We saw this today. Turns out Postini still used ORDB as of earlier today. This was reported (hostname changed a bit for privacy):
... while talking to postinicustomer.foo.s7a1.psmtp.com.:
----- Transcript of session follows -----
>>> RCPT To:<user@postinicustomer.foo>
<<< 550 64.18.2.63 blacklisted at relays.ordb.org
One would think that they OF ALL PEOPLE would know better!
Larry
Under the highly regulated AT&T regime that provided legendary reliablity, equipment was not cheap, and capacity was not plentiful. I'll wager you are way too young to remember. AT&T used to provide the phones. It was illegal to attach your own to the network. You couldn't go to the store and buy one. You couldn't get cordless phones. Only once that rule was changed did we start to see innovation in phones -- AT&T had no interest in giving you more than Princess. Likewise telecommunications. When I was young I had a 300 baud modem. Long distance rates were expensive... the BBSes/hosts of the day used a cell-network like method (recall Fido? UUCP? bang addressing?) to get email across the country. It took hours or more. Cell phones -- originally highly regulated and expensive (my Dad even had a wireless phone, precursor to the cellular network), with a single competitor to the incumbent telco permitted per market.
How did we get cheap cell phones, and a plethora of variety and capabilities? By competition, starting with the PCS spectrum auctions that ended the A-B carrier regime. How did we get the Internet? We got it when the federal government contracted with private companies to provide the backbone for the NSFnet, formerly the ARPAnet, and then began permitting interconnections. We got it through competition and lack of strict regulation. How did network equipment get so cheap and plentiful, so much fibre laid? You guessed it. Do you really think that if we had not had the environment created by that, the goldrush of the laissez faire Internet in the 90s, that networking equipment would be so cheap, that there would be such innovation, or that those "other countries" would have been able to ride on the coat-tails of that?
Fast, cheap, good. I've worked in the super-computing industry, for Motorola and some other telecom companies after that, and for several networking/Internet based companies before, during, and now after the boom. Trust me, it has always applied to far more than just manufacturing widgets.
Larry
Note what you wrote:
"Comcast and the rest can provide almost no service at all for what we pay them and they get away with it."
Note, "what we pay them." We pay them prices based on competitive forces, where reliability is just one factor. Granted, Comcast may not be the best example. But think in general.
The way the phone network got so reliable was because we granted a monopoly, and granted guaranteed, predictable profits. If it cost X to get the standards required, fine -- it was paid for, and there were *always* profits on top. That is key. We cannot have our cake and eat it too. Fast, good, cheap. We all know it. Look at the power companies. We have politicians pushing populist agendas to freeze rates, limit profits -- and the result is that the private companies running the power grid simply cut back elsewhere, and we have power outages and very little new power generation (of course that is also to be blamed on NIMBY opponents).
Perhaps we should have some more oversight of the Internet today, some sort of oversight board. But even if we did have that, would it or could it prevent the peering spats? Should it? What would a review board do in the case of an American company and European company with a contract that wasn't being honored? Would we need some sort of United Nations entity? How would this impact innovation and interest if decisions had to be brought before regulatory entities, subject to public comment, so on. If companies simply cannot depeer and make it actually hurt (what Cogent did by blocking traffic), then where do the incentives come from to provide the peering agreed upon? If we regulate it and mandate reliability, will we also regulate and mandate paying the true cost, along with a healthy enough profit to make sure a private entity remains interested in maintaining the network and providing for future growth and capacity? What will lack of competition do to cost, and market penetration? Will regulation drive away the private investors who fund these companies in hope of turning a healthy profit? Will we all pay for it via higher service fees or taxes? Is it perhaps ok to have these occasional spats, if the end result is a reasonably robust network at the "best" price? Or should the whole thing just be one big government funded and controlled system paid for by taxes and usage fees? Do we trust the government? Where would the innovation come from?
Perhaps what we have is the best of many imperfect possibilities?
Larry
Um, I used to use this Internet thingy when it was ARPAnet, before the advent of private backbones. I remember HOSTS.TXT and the real InterNIC. And yes, it was originally designed to route around major failures. That was one of the reasons DARPA, e.g. the military, funded it. It may have not done it perfectly, it may not have been able to survive a full-on nuclear conflict, but it was certainly designed and funded in good part as a research project into network robustness in the face of catastrophe.
Ever since the backbones went private though, all bets are off. You are entirely correct as of the early 90s. As we all know, it's "my network, my rules." Hence this peering spat, and the ones before, and the ones to come.
Larry
> I love it how Randians try to blame corporate corruption on governments.
I don't know whether I'm a "Randian" but I know that corporations are artificial legal entities that exist only by the consent of the government, legal entities that are essential. Those who rail against the existence of corporations, using the word "corporate" as an epithet, are no better than those on the other end of the spectrum. Corporations are artificial entities -- they cannot be evil, only people can be evil.
Government is certainly partially culpable for the worst corporate excess when the environment for that excess was greased into existence by money finding its way into the hands of corrupt politicians of all political bents.
In your extreme example of a monopoly on food production and distribution, the only way that could happen is if government did not work properly in its oversight of that legal construct. It isn't the corporation that is evil, it is the people involved in the little drama who are evil. The people who run the corporation, the people who work for it doing evil things, the shareholders who learn of the evilness and do nothing, the police paid to not intervene, inspectors, judges, politicians, so forth. I guarantee that all of those would be involved in the creation of such a scenario. They are the evil ones. And ultimately the people themselves, because what else is "government" than a collection of people, for permitting that environment to exist. To simply rail against "evil corporations" is childish. The road was well traveled.
So I don't think the original premise is entirely correct -- the biggest, nastiest corporation can *force* you, but only after it *enticed* a lot of people, including those in government, to permit it to become big and nasty. So where does the blame fall? And who is evil?
Larry
After this debacle, Democrats need to figure out what their goal is. While their system may be "fair" it is also fairly stupid. The party needs to get a candidate and not excessively damage that candidate in the process of choosing him/her/it. They should either have all states vote in a very shortened time span, or move to a winner take all system. Instead, this thing will drag on due to their short-sighted but "fair" proportional allocation of delegates. That may be fine in a real election, done on one day. But not like this, dragged out over months. It's idiocy.
Larry
Not to mention: why is the taxpayer footing the bill for any portion of the primary process? These political parties are private clubs. How they choose their candidates should be their business, using their rules, and on their dime. Not only should the law not mandate "open" primaries, the law should not mandate caucus versus secret ballots, nor treat these any differently than your local Bingo club electing officials. The election is in November.
Larry
Your anecdotal experience is one thing, but facts are facts. Having said that, this is much like a LP vs CD vs MP3 discussion.
I worked for Motorola in the Cellular Infrastructure Group back in the mid 90s, the period when PCS license blocks were being auctioned off by the FCC and we began the transition from a dual-carrier marketplace to our modern multi-carrier marketplace in the US -- I later went to work at one of those PCS license-holders for a bit. I'm a bit dated on some things in the industry, but the fundamentals at the transmission level haven't changed all that much. AMPS exclusively dedicates a fixed amount of bandwidth from frequency A to B to your voice, an amount derived from the channelization of POTS lines. That was FDMA -- the total allocated bandwidth was split into two full dedicated audio channels per conversation, with no overlap, chosen primarily to match POTS voice bandwidth and quality. N-AMPS came later and sacrificed some bandwidth, but still provided plenty for the voice. TDMA is digital sampling and packetization of the voice, sorta like VOIP these days. The primary design criteria for TDMA was to increase bandwidth utilization with an acceptable loss of voice quality, along with providing additional features. When I was at Motorola there were guys working in a lab nearby with Qualcomm on the early CDMA technology -- it was in theory capable of providing voice quality on par with AMPS, but like TDMA one of its primary goals is utilization with acceptable quality degradation. It didn't work so great in those early days, but is now a superior technology for voice quality I believe. In fact next generation GSM is an amalgam with CDMA mixed in.
That's just the raw technology though. Implementation is key. All things being equal, we are not likely to get better pure voice quality on a consumer cellular network than was had back in the days of AMPS with analog engineers pursuing voice quality as job one, not capacity. There are too many cell phones now for that to be the goal.
Larry
That TDMA thing always irks me. Current GSM is TDMA, you know (though next generation will be CDMA). It means Time Division Multiple Access, essentially packetizing data. So while some forms of TDMA may be obsolete, the world's most popular mobile phone system also is a form of TDMA.
Larry
I'm sorry, but better sound? AMPS with a clear signal is the best you are going to get. AMPS, a form of FDMA, was designed back in the days of just an A carrier and B carrier per market, when cell service was expensive. AMPS, and even N-AMPS, wasn't designed to maximize calls per antenna, it was designed to match the voice bandwidth used by analog telephone lines, essentially a wireless equivalent. Everything since then has been designed to maximize capacity at the expense of voice quality. PCS, TDMA, GSM (which is TDMA BTW -- TDMA is not analog), CDMA, all have as a major design criteria increasing the call capacity of limited frequency bandwidth. The result is greater competition, reduced cost, wider availability of cell phones, everybody and their pet carrying one, etc., all good things perhaps -- but "better" sound quality? I don't think so.
Larry
Good point. But it's part of what I consider the order of things to do. My obligation as a buyer is to pay for the item promptly or as otherwise agreed. My obligation as a seller is to sell the item I described and ship it as agreed. Once the buyer fulfills their side, the seller should say so. Once the seller fulfills their side, the buyer should say so. Somebody has to go first, unless your idea of blind feedback, which I do like quite a bit, were implemented. I feel it is unreasonable that the buyer leaves feedback first, but is fearful of leaving honest negative feedback for fear of an unwarranted retaliatory negative response. Sellers tend to have more leverage in the system than buyers, particularly in this day and age where so many sellers are businesses/quasi-businesses.
And yes, whatever. I've always done it this way, it's my policy (even published on that About Me page). Sometimes as a buyer I never get feedback, because the seller never leaves it, even when I do tell them that all is well and I'm happy. I've been told by many that they have an automated feedback system that leaves positive feedback only once I do. Ridiculous. And yes I run some risk of getting a negative feedback after I, as seller, already left mine once the buyer paid as agreed -- but that's the risk you run. I'm honest, I charge fairly for shipping, and I have 100% positive feedback over 10 years, and have never had a bad ebay experience. Well, I take that back. Two not so perfect experiences, both involving bubblewrap and insufficient peanuts. In both cases the sellers offered to make good for the damage.
Larry
I disagree. As a seller, I leave feedback once the person pays, as a rule. That is their obligation to me. They pay, I ship. If they complain via feedback, I can respond. But I never, ever hold feedback hostage. And likewise, as a buyer, I never, ever leave feedback until the seller does.
And I certainly have noticed in recent years how many sellers do not leave feedback until they see what you have to say. Unfortunate. But I'm iron-clad in my policy on feedback, I've been doing it that way for ten years now.
Also, anybody remember when you could put comments on an active auction, e.g. "are you crazy, you buy that widget for $X brand new in the store!" I hope they bring that back. Ebay has gotten way too seller/merchant biased.
Larry
No, there are lots of cars with transverse engines that are home-mechanic friendly. My Dad's a Buy American kind of guy with cars, and his cars have always been more difficult to work on that my Japanese cars, mainly Nissan and Mazda. Seems most American cars expect you to be using a mechanic with a lift for everything. I remember on one of my Dad's American cars, trying to change his oil one time with a floor jack in the driveway... it wound up being easier to take off a wheel to get access to the filter than do it from below on the ground.
Larry
Regarding the add to cart -- no, that is not accurate. I just searched for a list of domains -- just searched -- and all are now held by NSI. However, it holds only the .com version, although I did not provide a TLD in my query list. I am so writing a program to automate this. It appears you have to feed it in through their web CGI, a whois query doesn't result in the hold.
Larry
OMG. Why did you have to do that? Now I remember using IE on a Sun workstation!
Larry
Yeah, I don't think you got the comment. Monetize a browser? The irony is precious given the history of IE.
While I agree with you in:
t ml
How about the one where the government can now IN SECRET, presenting no evidence or holding no hearing or trial as GUARANTEED by the Constitution, seize your assets? How is that "protecting the constitution?
This does seem a rather natural progression given what we've been doing to our rights for A LONG TIME. Ever heard of civil forfeiture? It's been in use for a long time before Bush.
http://www.law.cornell.edu/supct/html/95-345.ZS.h
Gotta love cases like United States v. One Assortment of 89 Firearms.
Bush may be doing things that are quite objectionable to quite many, but the funny thing is, things that would have been clearly unconstitutional in the past (such as, oh, a federal minimum wage, or federal workplace safety laws) suddenly become constitutional when expedient, with no substantive changes in the underlying text or intent. I always tell people, a constitution subject to interpretational whim is a slippery slope. If you let politicians and judges decide how our fundamental legal foundations change without the bother of amendments and the entire associated process, then odds are it will change in ways you both like and do not like. You can't have it only your way. Unless you try to pack the Court. Like FDR.
Larry
Fire trucks are fun. My eccentric dad has a '55 Mack pumper, and a '40-something 100' open cab American LaFrance ladder. I've run a red light or two... it's a pain to get that pumper going, straight cut gears and all. We need to get them running again this summer.
Larry