I have been using a Lucent WaveLAN 802.11 adapter hooked up to a 1 watt amplifier and 24db antenna to hook up to the Internet through my local ISP for the last year.
The good:
1. Wireless is easy to install. If you are close enough to your nearest access point, just pop in a pcmcia card with a pigtail, install drivers, and you're done.
2. Wireless is extremely reliable. The link itself basically never goes down. The only times I've had a bad signal to noise ratio are when I screwed up my internal wiring to my antenna and during very high winds.
3. Wireless is very fast. At up to 11Mb/sec, it's one of the fastest access methods available in the price range.
4. Wireless is cheap for the ISP. Initial setup cost for your ISP is lower than some comparable technologies (DSL particularly). They can hook up 30 customers to each access point using the same frequency-hopping spectrum, and add cheap additional hardware for each new group of 30 customers.
5. Wireless is cross-platform. Many drivers are free software.
The bad:
1. Wireless can be very complicated to troubleshoot. It's easy for the customer to screw up their link; when something flakes out, it is often tough to tell whether it is you or your provider.
2. Wireless is very expensive for the consumer. While setup costs for people very close to an access node is relatively small (less than $250 for the card and pigtail), costs range up to $1000 for people further away.
3. Wireless can be very slow. 802.11 is designed to slow the link as the signal to noise ratio drops. You may connect as slowly as 64Kb/sec.
4. Wireless can be tough to install. Setting up my link involved several hours of attic and drill time.
5. Wireless competes in common frequency ranges, and has the usual problems with radio transmissions. As the so-called "medical band" (2.4GHz) becomes more cluttered, you're going to notice higher packet loss and latency, conflicts with cordless phones, etc. Also, hills, trees, and bridges can all interfere with your line-of-sight to your ISP's antenna.
6. Wireless has trouble scaling over distance. Your ISP will need a repeater every mile or three in order to broaden their service. DSL and cable have other, similar costs, such as upgrading local loops.
My take? I believe wireless is a great technology, and will continue to enjoy a strong and growing *large* niche market. I know I love mine : )
Just a few years ago, the Boy Scouts of America managed to prohibit homosexuals from being scoutmasters, and now they are into software piracy investigations! Will wonders never cease...
Matt Barnson
Re:I think there are some things to be cleared up
on
Is Novell Doomed?
·
· Score: 2
Clarification in order here:
NDS *predated* LDAP.
NDS evolved because of X.500 and DAP (Directory Access Protol)'s weaknesses (principally, the reliance on a tiered directory request system that didn't work well.) Novell replaced X.500's proxy request system for authentication with a "referral" system -- allowing one NDS server to refer a client to another NDS server's resources without making the request itself.
The standards bodies took some of the best ideas from DAP and NDS, revised them to suit their ideas of how it should work, and created "DAP Light", or LDAP.
So LDAP came about because of NDS -- not the other way around. Now NDS supports LDAP, too, and everyone benefits.
I've read most of the higher-rated comments here, and they have missed an ENORMOUS advantage Novell has in the ISP/ASP frontier:
Their Internet Caching System.
Novell figured out several years ago that their BorderManager server scaled better on static page serving via its caching services than most web servers did. They placed two BorderManager servers in a reverse-proxy setup in *front* of their own web servers, and watched their web servers be able to serve up DRAMATICALLY more pages than they could alone.
They created a development effort to strip down BorderManager, rebuild the file system into a BTREE, and parlay this caching service into an integrated, vendor-only hardware/software solution to leverage into a platform. They have largely succeeded.
We have two Compaq boxes running Novell's Internet Caching System. ICS is not something you can just buy off the shelf. It's heavily engineered for the systems on which it runs. Let me share some statistics.
One of our web sites has, historically, run over 4 million hits per day (www.excitestores.com, if you're interested). ICS reduced our number of Apache processes running from about 30-60 at any given time to *7* (that's 5 base HTTPD processes, plus two). The memory load dropped tremendously, and ICS achieved a 94% page hit ration in this reverse-proxy setup, with a 57% byte hit ratio. This means that our e-commerce setup, which heretofore required 10 web servers to service, now only requires TWO, plus two ICS boxes. And the load on the ICS boxes has never passed 5% CPU utilization.
ICS also does some other pretty amazing things. It can "SSL-ize" content, so you store your certificates on the cache server and you don't need to compile mod_ssl for Apache (which, in case you didn't know, is a HUGE hit on your CPU). You can arrange content delivery to remote sites (ala Akamai's service) and have enormous bandwidth savings. You can leverage the ability of the box to intelligently handle your traffic and scale to over 100,000 simultaneous connections (Our Apache daemons, at about 10 MB each on a box with 2GB of memory, are far more limited than that).
Suffice to say, I believe that Novell has a long and prosperous future ahead of it. We evaluated many different caching technologies, and chose ICS over Squid and several proprietary solutions; even in our UNIX-savvy environment (all our production services run on UNIX except the caches), ICS won out. Novell has already learned the lesson Linux is starting to: success comes, not necessarily in being the best at what you do, but being darn good at an awful lot of things. Their recent staffing cutbacks reflect the changes in strategy from a "software company" to a "solutions company".
DISCLAIMERS:
I do not work for Novell. I have just purchased their products because of technical superiority. Novell is fighting an uphill battle to get into a company dominated by Solaris and Linux. So don't assume I'm a troll because I like what I see : )
In the past few years, the United States Patent & Trademark Office (USPTO) has taken a very aggressive position in approving patents on business practices and software algorithms. Taxpayer subsidies of this office have also been largely eliminated, requiring the Office to "pay their own way".
This has led to a slew of patents that should never have been granted, and never would have had the patent reviewer been familiar with the technology involved. For instance, the Amazon.com "1-click" patent is often referenced as an example of a business practice that should never have been patented. In case you were not aware, Amazon.com claims this patent covers all Internet business methods allowing customers to use "One Click" to purchase goods and services over the internet, using techniques that are obvious to those who understand how to create interactive web sites.
This is one example of a lack of quality control standards and proper understanding on the part of the USPTO regarding technology and the Internet. There are many more being awarded daily, as you can find out for yourself by doing a search for "patent" at http://www.slashdot.org/search.pl.
If elected, what are your plans, if any, for reform of the USPTO to prevent further abuse of our patent system? If you believe they are currently doing a good job and should not be reformed, how do you defend this position?
You can find many of these patent problems the subject of past slashdot discussions:
It's really nice to see that we have people reacting to the DeCSS fiasco in a positive, intelligent way. For once, instead of posting anonymously to a Slashdot forum, people are takinga action.
HOORAY!
Now, if only the Your Rights Online section had the readership of the rest of Slashdot, an article like this might deserve a fair shake.
I am in the process of trying to find entry-level Systems Administration staff for my company right now; let me give you an insight to life on the other side of the fence.
First, about myself: I'm 27, I've been working mainly in the Internet industry since I was 21. College dropout because work was too interesting and my grades stunk. I was hired at my current company as a mid-level UNIX Systems Administrator, with two years UNIX experience (and five years Windows/DOS/Macintosh experience). I am now the manager of Systems Administration for Excite Business Applications in Utah.
I'll freely admit that I see age discrimination regularly. Many managers assume that older IT workers do not have the current skill sets to handle our work. I believe that's wrong, but having been in the business six years now I'm noticing that the average age is going up -- whereas most workers were under thirty before, I'm regularly seeing people into their forties and fifties in positions of responsbility at my company and elsewhere. I really feel that it's the *industry* which is finally growing up. There is an enormous shortage of competent, skilled workers in the field. From anecdotal evidence, I would suggest somewhere between five out of ten and nine out of ten applicants do not have any clue about their jobs. That's pretty sad, and makes us (managers) very jaded.
I'm looking for bright, competent workers in our entry-level systems administration positions. We hire more experienced admins as well; please allow me to offer a few tips on resumes that really make the difference for me on whether I ask someone to interview with me:
Do not list time frames for previous positions on your resume. This avoids potential pre-interview age discrimination. Most employers aren't actively discriminating against older people, they just assume you're far too expensive for the position. Don't let them know your age on the resume.
Remember, your resume is an advertisement about yourself, not a listing of your previous failures. Talk yourself up!
List only relevant experience about yourself in your resume. If a job you held 15 years ago is relevant, list it. If the last job you held had nothing to do with the position for which you're interviewing, don't list it.
The only way I'll believe your Summary of Qualifications is if I see work history that backs up each bullet point.
Leverage your age as an asset in the interview. Convince the interviewer you are reasonably priced (their first concern when interviewing someone older than they are), more competent than any "young gun", and (most importantly!) able to learn quickly.
Many employers assume older workers are high priced, out-of date, and slower to learn than younger people. It's the prospective employee's job to recognize age discrimination exists; prove to the interviewer that these stereotypes do not apply to you. I hate to say it that way, but there are a lot of clueless newbie managers out there who simply don't know it's illegal, unethical, and stupid to discrimate based upon age.
The law says all employment in the U.S. is "at will" -- you can leave, be terminated, or not get a job for any reason or no reason, outside of the following federally-protected exceptions:
You cannot discriminate based on age, if the employee is 40 or over. You can, however, discriminate against people 39 and younger based on their age, legally.
You cannot discriminate based on race or sex
You cannot discriminate based on sexual orientation in the state of California. Most other places it is legal, but unethical (and if they sued they'd probably win)
You cannot discriminate based on marital status
You cannot discriminate based on any physical or mental disabilities. If the disability would interfere with the primary job function, you can discriminate (ergo: a person with no hands cannot be a hand model), but if the disability can be reasonably worked around, you cannot discriminate based upon it(ergo: a person has no legs and applies for a bicycle courier position. The disability can be worked around by providing them a hand-powered bicycle).
I believe there are more, but cannot recall right now.
These are the things I, as an employer, willfully discriminate upon:
Experience
Work ethic
Formal training (school)
Ability to work the given hours
Ability to teach oneself (this may be in conflict with the ADA (Americans with Disabilities Act), since I require highly-intelligent people in my organization; someone with a severe learning disability would have trouble in this position. Should a severely learning-disabled person with Systems Administration experience apply for my position, I'll be in a tough spot)
Ability to work unsupervised. Lots of people can pretend to work when unsupervised.
Linux home hobbyist experience
GNU-friendliness. I want someone who likes the GNU GPL and will work his butt off to make sure we're in compliance at all times.
I wonder if I'm the exception rather than the rule? I certainly hope someone out there was benefitted by this post; if not, at the very least it helped me clarify my thoughts on this issue.
There's a very simple technological solution to your problem: caller ID. Most decent caller ID boxes have a mode where you refuse to accept calls from people who do not allow their phones to be identified with caller ID. Most telephone solicitors have numbers that do not allow them to be ID'd, and problem solved. Those who do, just glance at your caller ID box, see that it says "SpammerzAreUzz" and refuse to answer the phone! We have it, we like it. However, it costs $$, and has the potential of locking friends who value their privacy out of your life via phone...
I'm writing this comment using a Lucent WaveLAN at this very moment. My setup closely resembles yours, except my WaveLAN connection is to my ISP The driver support under Linux for the WaveLAN card is fantastic, and the driver is even included with Linux-Mandrake 7.1 -- no special configuration required other than to make sure pcmcia inits before network. The downside? It's not a 10-Mbps connection, it's only 2-Mbps. However, I routinely run X connections over SSH from a dedicated ssh gateway at my business. The link has fantastic uptime, the signal-noise ratio is almost as good as it gets with a 24dB antenna, and I'm completely undisturbed by power outages (UPS here), snow, rain, etc. High winds (60 m.p.h. + gusts outside right now) do take a toll, though. A 1 watt amplifier allows my 24dB antenna to punch right through the walls of my attic and lets me forget about the many trees, houses, and other obstructions between me and my ISP, 3 miles away. The WaveLAN card with a pair of antennas and amps would be a much easier, more convenient option for most home users trying to set up this type of connection, IMHO. The amps are pretty expensive (a couple hundred dollars) but do wonders to ensure connection integrity. Here's what my S/N ratio looks like, catting/proc/net/wireless with a gusts up to 60 m.p.h. and sustained winds at 30-40 m.p.h right now:
Inter-| sta-| Quality | Discarded packets
face | tus | link level noise | nwid crypt misc wvlan0: 0000 14. 179. 164. 0 0 8612
Normally the link level is 30-32. There is a steady falloff in bandwidth as your signal to noise ratio drops; right now I'm probably down to a 512K or 256K link because of the storm.
Basically, reverse proxy caching works by you hijacking connections to your webservers from the outside world. IP Policy-based routing is the easiest example to understand, and is the method we use at Excite@Home E-Business, so I will detail it. A connection is destined for "www.excitestores.com", and ends up at the external DS/3 (T3, T1, insert your fast link here) port on our router. The router runs a rule against the packet and says "Hey, this is www traffic bound for the servers that are to be accelerated. Therefore my next hop is (insert IP address of cache here)!". It route-maps it to the cache server as it's next hop. The caching server is set up to "hijack" any incoming connections as if they are destined for itself, and makes the request to the origin web server on behalf of the requesting client. At this point, this does not differ too much from standard forward transparent proxying, except that you normally have an access control list that only permits transparent proxying of a limited set of URL's or IP addresses. You don't want to run an "open proxy" for the world to use to cache whatever they want. Of course, note here that there are alternate methods of accelerating sites depending on the cache you choose and your infrastructure. The basic idea is to get the packets to your cache instead of the web server, however you choose to do it. Common methods include placing the cache in the natural route of the packets, making the webserver address point to the cache and have a non-public DNS that the cache looks to to resolve a web site on a non-routeable private network, or specifying on the cache that incoming connections on a certain IP are to accelerate a particular origin server. Anyway, the benefits of this are enormous in our case. We have a (*&$load of modules compiled into our Apache server, tons of virtual hosts and modules to handle them all, and each daemon runs about 12 MB. Each web server has a gigabyte of RAM, therefore you do the math: 1024/12=85 and 1/3 connections run us out of physical RAM on each web server. Realize this is a rough estimate; our web servers can handle much more, but performance degrades quickly with more connections being served from virtual memory. I've also not taken into account OS overhead, other services running on the servers, and any other thing you may think of. However, modem users, particularly, saturate web server connections because it is so slow to deliver objects to them. CNN.com, for instance, uses ICS caching boxes purely for connection management to handle these slower connections that could bog their servers down. Novell's ICS is rated at over 100,000 simultaneous connections on each box in reverse proxy mode. A big difference from 85 connections for one machine, no? I'd love to discuss this in more depth, if you require a better answer. Better yet, check the FAQ at Squid's site regarding transparent reverse proxying.
Seriously, this is what takes web sites to the next level, regardless of whether you use Squid, ICS, NetCache, or another type of reverse cache. Keep smiling!
I've been overseeing a caching (really, website acceleration project) for my company, Excite@Home E-Business Services, over the last three months now. I can personally say that the three I've had experience with, Novell's ICS caches (which comprised ten of the twenty entrants), Network Appliance's NetCache, and Squid (on Solaris, in our case) all rock. Squid 2.3-stable1 was a dream to compile, install, and configure. ICS has a few user interface quirks with their Java administration tool that I don't like, but except for Cisco's cache (Oh My Gosh do you really want to spend $150,000 on a CACHE???) ICS-based systems captured the many top honors in this roundup. Network Appliance's NetCache is also a nice choice, and as the only vendor with streaming media caching/splitting support, they are receiving a lot of attention recently. It's really important to note that IRCache has no desire to point to any "winner" in this bakeoff, but instead to have real non-partisan numbers to point to when evaluating cache performance. Squid captured top honors in cache hit ratios, but nothing else (AFAICT), showing that those "expensive, proprietary systems" also can be very well-tuned operating systems that eliminate traditional OS overhead for these numbers. One of the frequently overlooked uses of cache is as a web site accelerator, instead of the standard forward proxy. Using a few simple access control lists and a policy on a router, reverse-proxy caches managed to reduce the instantaneous load on our web servers by up to 94%. We serve about 3.5 million hits a day. A "reverse proxy" is an EXCELLENT use of a proxy cache, and after these technology evaluations I've been involved with in past weeks I'd recommend it to anybody considering running a high-traffic website. This allows your Apache servers to function more as the "cgi engine" of your site, and lets the static images, text, banners, etc. be delivered from a box that can handle 100,000 simultaneous connections. Very cool. While I'm not allowed to post a "review" of any one of these units, because of various agreements for the evaluation boxes we tested, I can clearly state that Squid, NetCache, and ICS-based systems can and will vastly reduce infrastructure scalability costs for businesses when deployed in a reverse-proxy configuration. Our earlier estimates guessed we'd need to expand our web farm three times to handle our estimated load by the end of the year. Now we can reliably predict that our farm can serve 10 times the amount of hits we're running now by using a cache as an accelerator. VERY cool stuff. Be sure and check out the system configurations in the bakeoff review. It's very illustrative that the boxes tested have VERY specific audiences. Don't be fooled by the "fastest hit response time" or "most throughput" -- you can spend $6,000 or $150,000 for any setup, depending on your needs. Noticeably absent from the review was Inktomi, for the second year in a row. I'm hearing FUD from vendors that their performance isn't up to snuff-- any truth to these rumors?
You know, we've run up against this same problem time and again in our new "connected" world. That the concepts of "taking" and of "property" lose meaning in the context of software. It appears our current judicial system just isn't well equipped to answer these questions. As with many things, I think it depends on one's point of view. From the author's point of view, if he views his algorithm as "property", it could be considered taking and he could demand payment. Alternately, since he's one who considers computer source code a form of speech (as do I), the restriction on exports of source code may be construed as a restriction on his first amendment rights. It's a complex issue that isn't going to be solved in a day. The original issue with this guy was that the export restrictions, which were never voted upon nor approved by representatives of the people, but instead were handed down by Bill Clinton, a) should not apply to personal speech, and b) were an unconstitutional executive order. I don't know what the answer is, but my representatives hear from me regularly on these kinds of issues. Remember that one consituent phone call is worth thousands of lobbyist dollars! Sorry to ramble. Out.
I work for iMALL, Inc. which is soon to be acquired by Excite@Home. We just met today with Joe Kraus, one of the six co-founders of Excite. One of the questions asked was "Is there any truth to the rumor that Excite@Home is considering merging with AOL?" The answer? "No. I have no idea where this current rumor started. Well, I have some idea, but it really looks like they've recirculated old news about negotiations that were stopped three months before the rumor hit the papers." He went on to say that yes, Excite@Home was talking with AOL months ago about putting portal links up on Excite pages for AOL/Netscape services. AOL thought Excite@Home's requirements were too high (a stock-sharing arrangement with AOL in exchange for the space) and the negotiations were dropped immediately (we're talking like 6 months ago). The current restructuring is necessary to ease the burden on corporate heads and to bring divisions more inline with Excite@Home's underlying corporate philosophies and "the way things work" in the company. Things have been a bit uncomfortable since Excite's merger with @Home, but now the two complementary businesses (broadband access & search/portal business) support each other better, as well as make room for the upcoming merger with iMALL to serve as their e-commerce division. I hope this has helped dispell this unfounded rumor.
Boy, one would think this guy knew his Marx and Lenin better... AFAIK, Communism is rooted in the idea that the working man would rise up and overthrow the ruling class, distributing the fruits of labor equally to fellow proletariat. This man thinks the free software culture is communism? Bah! Humbug! The Internet is STILL a tiny club of culturally-elite, rich (compared to the rest of the world) burgoise representing everything despised by Marx. Despite our burgeoning population, we represent a tiny fraction of priveleged humanity trodding upon the backs of the repressed masses. We distribute our gifts to other members of this elite class of intellectuals. The Free Software culture is far from Communism. It more resembles Nepotism, with the talented, rich few giving away to the less-talented rich few -- yet all members of the same, elite club. Maybe if we could bring the Internet to the huddled masses with no concept of computers... But we can't even feed most of them. I must agree with the comments noted in the review from others: The idea that the "gift culture" is communism is hogwash.
I've hand many chances to play with machines with redundant power supplies. In two words, THEY SUCK! Failure rate on redundant power supplies is ENORMOUS. In most cases (double entendre intended) two power supplies must occupy the space normally inhabited by a single power supply. They fail an order of magnitude more often than large single power supplies. IMHO, your overall best bet is redundant servers. Linux doesn't have many built-in high-availability features but the High-Availability HOWTO is a great place to start. But redundant power supplies -- PHEW! They stink!
Many of the comments I've read are quite insightful, but do not seem to address the specific problem of donating time and/or resources to Salt Lake City-area school district computer departments. I am a network engineer for the Granite School District, the largest in the Salt Lake City area, and feel I am qualified to answer your question.
We are deploying several hundred Pentium 2-class PCs throughout the district over the summer with the goal of every single classroom having an Internet-capable computer at their disposal. In addition, the District has just approved $3 million dollars for new hardware purchases so that each teacher may have a reasonable machine upon which to work and browse the Internet. Internet access is filtered (at least HTTP access) through the Utah Education Network to remove objectionable material. Internet access is provided at the head-end through a single T1. So, to answer your questions:
Unfortunately, I am unqualified to answer questions about tax issues for nonprofit organizations. However, the Computers for Schools program routinely donates scores of computers to Utah schools. Individuals and businesses donate PCs to this program using the process outlined at the above URL, which are then cleaned up, and in some cases upgraded (To a Cyrix M2/GX 300 w/32 MB RAM and 4GB HDD) by prisoners in Utah detention centers, shipped to schools that have requested them, and put to use. BTW: The systems we get after the prisoners have worked on them are CLEAN, and I've never seen a single upgraded system go bad yet. They really do a good job -- hurrah for rehabilitation!
You would likely be rebuffed if you attempted to administer, repair, or install systems at schools (although just dropping off a network-and-Windows 95 capable machine would net you many thanks). A dozen qualified Network Engineers oversee 89 SysOps to make sure machines are installed, maintained, and correctly configured. There are strict security controls on all lab computers which generally only run Corel WordPerfect, NetScape Navigator, a couple security packages, and whatever specialty software is needed (scanning software, printer drivers, educational games, etc.) Your time would be very well served by volunteering to teach classes, run extracurricular clubs, or mentoring children (and teachers!) in PC technology. The administrators are generally receptive to someone willing to help or supervise children using computers before, after, or during schools. It is up to each principal if s/he wishes you to assist, however. Additionally, you must be careful (as one poster mentioned) that you do not become a commodity. You must set strict hours and limits on your participation, else you will be used to exhaustion.
If you really, really want to help fix systems, not just mentor children in their use, in Salt Lake City area schools, contact the Granite School DistrictHuman Resources Department and let them know you wish to volunteer -- they'll definitely be able to help you out.
I was one of the unwashed masses who were unfamiliar with Linux just a couple years ago. I was firmly in the Microsoft camp, doing daily drone drudgery to maintain PCs for customers. One day, while trying to find a cheap webserver, I discovered Linux and fell in love with it. Does it feel good to be admired by your peers for competence with some "powerful, obscure" operating system? Absolutely! Is it fun to make comments in meetings such as "to do that would be trivial with Linux"? Of course! Does the fact that nobody else at work knows my chosen server and workstation operating system make me feel unique and special? Hell, yes! From my experience: *Knowing Linux and UNIX in general pays more than NT/NetWare expertise *My knowledge of Linux makes me irreplaceable in the eyes co-workers and employers *Linux does what I want it to do, the way I want to do it, for as long as I need it to, without crashing
When the Internet first went commercial, folks who knew HTML were paid big money. When people realized their applications weren't Y2K compliant, Cobol came back (briefly) into vogue. When the day comes that I get paid more for knowing Amiga:), BeOS, or another OS, and Linux expertise is as common as donuts, I may consider it time to acquire some new "specialized skills" to make more money and be regarded as a "computer guru" with "the latest thing".
To sum up: I agree with the author, except that I think that there must, and will, always be enough new developers learning to work on Linux to supplant those who choose to move on. If not, Linux would become a footnote in the dusty tomes of history, with the UNIVAC, OS/2, and Charles Babbage's "Analytical Engine" (ca 1837).
I'm afraid you're mistaken: CD Changers do have an important place in the enterprise -- but probably not ones this small.
In the right circumstances, a CD changer can be an important, vital storage device. Large CD Jukeboxes, like the types made at Meridian Data are perfect for long-term, near-line storage of critical data. Many models serve over 500 CD's (that's well over 300GB, quite an expensive set of hard drives) with anywhere from 2 to 8 CD-ROM drives allowing for multiple simultaneous access of different CDs. You can also get jukeboxes with CD-R drives for permanently burning important data to a media that has a shelf life of up to 50 years (better than an average hard drive's 10 years).
CD Changers and Jukeboxes are a vital part of any comprehensive Off-line/Near-line/On-line backup strategy, along with tape archives and hard drives. But laborious load times, hard user limits, and lamentable throughput will probably dog the media forever.
However, in the right situation it is the perfect solution.
Slashdot Mirror
The good:
1. Wireless is easy to install. If you are close enough to your nearest access point, just pop in a pcmcia card with a pigtail, install drivers, and you're done.
2. Wireless is extremely reliable. The link itself basically never goes down. The only times I've had a bad signal to noise ratio are when I screwed up my internal wiring to my antenna and during very high winds.
3. Wireless is very fast. At up to 11Mb/sec, it's one of the fastest access methods available in the price range.
4. Wireless is cheap for the ISP. Initial setup cost for your ISP is lower than some comparable technologies (DSL particularly). They can hook up 30 customers to each access point using the same frequency-hopping spectrum, and add cheap additional hardware for each new group of 30 customers.
5. Wireless is cross-platform. Many drivers are free software.
The bad:
1. Wireless can be very complicated to troubleshoot. It's easy for the customer to screw up their link; when something flakes out, it is often tough to tell whether it is you or your provider.
2. Wireless is very expensive for the consumer. While setup costs for people very close to an access node is relatively small (less than $250 for the card and pigtail), costs range up to $1000 for people further away.
3. Wireless can be very slow. 802.11 is designed to slow the link as the signal to noise ratio drops. You may connect as slowly as 64Kb/sec.
4. Wireless can be tough to install. Setting up my link involved several hours of attic and drill time.
5. Wireless competes in common frequency ranges, and has the usual problems with radio transmissions. As the so-called "medical band" (2.4GHz) becomes more cluttered, you're going to notice higher packet loss and latency, conflicts with cordless phones, etc. Also, hills, trees, and bridges can all interfere with your line-of-sight to your ISP's antenna.
6. Wireless has trouble scaling over distance. Your ISP will need a repeater every mile or three in order to broaden their service. DSL and cable have other, similar costs, such as upgrading local loops.
My take? I believe wireless is a great technology, and will continue to enjoy a strong and growing *large* niche market. I know I love mine : )
Matt Barnson
Just a few years ago, the Boy Scouts of America managed to prohibit homosexuals from being scoutmasters, and now they are into software piracy investigations! Will wonders never cease...
Matt Barnson
NDS *predated* LDAP.
NDS evolved because of X.500 and DAP (Directory Access Protol)'s weaknesses (principally, the reliance on a tiered directory request system that didn't work well.) Novell replaced X.500's proxy request system for authentication with a "referral" system -- allowing one NDS server to refer a client to another NDS server's resources without making the request itself.
The standards bodies took some of the best ideas from DAP and NDS, revised them to suit their ideas of how it should work, and created "DAP Light", or LDAP.
So LDAP came about because of NDS -- not the other way around. Now NDS supports LDAP, too, and everyone benefits.
Matt Barnson
Their Internet Caching System.
Novell figured out several years ago that their BorderManager server scaled better on static page serving via its caching services than most web servers did. They placed two BorderManager servers in a reverse-proxy setup in *front* of their own web servers, and watched their web servers be able to serve up DRAMATICALLY more pages than they could alone.
They created a development effort to strip down BorderManager, rebuild the file system into a BTREE, and parlay this caching service into an integrated, vendor-only hardware/software solution to leverage into a platform. They have largely succeeded.
We have two Compaq boxes running Novell's Internet Caching System. ICS is not something you can just buy off the shelf. It's heavily engineered for the systems on which it runs. Let me share some statistics.
One of our web sites has, historically, run over 4 million hits per day (www.excitestores.com, if you're interested). ICS reduced our number of Apache processes running from about 30-60 at any given time to *7* (that's 5 base HTTPD processes, plus two). The memory load dropped tremendously, and ICS achieved a 94% page hit ration in this reverse-proxy setup, with a 57% byte hit ratio. This means that our e-commerce setup, which heretofore required 10 web servers to service, now only requires TWO, plus two ICS boxes. And the load on the ICS boxes has never passed 5% CPU utilization.
ICS also does some other pretty amazing things. It can "SSL-ize" content, so you store your certificates on the cache server and you don't need to compile mod_ssl for Apache (which, in case you didn't know, is a HUGE hit on your CPU). You can arrange content delivery to remote sites (ala Akamai's service) and have enormous bandwidth savings. You can leverage the ability of the box to intelligently handle your traffic and scale to over 100,000 simultaneous connections (Our Apache daemons, at about 10 MB each on a box with 2GB of memory, are far more limited than that).
Suffice to say, I believe that Novell has a long and prosperous future ahead of it. We evaluated many different caching technologies, and chose ICS over Squid and several proprietary solutions; even in our UNIX-savvy environment (all our production services run on UNIX except the caches), ICS won out. Novell has already learned the lesson Linux is starting to: success comes, not necessarily in being the best at what you do, but being darn good at an awful lot of things. Their recent staffing cutbacks reflect the changes in strategy from a "software company" to a "solutions company".
DISCLAIMERS:
I do not work for Novell. I have just purchased their products because of technical superiority. Novell is fighting an uphill battle to get into a company dominated by Solaris and Linux. So don't assume I'm a troll because I like what I see : )
Matt Barnson
In the past few years, the United States Patent & Trademark Office (USPTO) has taken a very aggressive position in approving patents on business practices and software algorithms. Taxpayer subsidies of this office have also been largely eliminated, requiring the Office to "pay their own way".
This has led to a slew of patents that should never have been granted, and never would have had the patent reviewer been familiar with the technology involved. For instance, the Amazon.com "1-click" patent is often referenced as an example of a business practice that should never have been patented. In case you were not aware, Amazon.com claims this patent covers all Internet business methods allowing customers to use "One Click" to purchase goods and services over the internet, using techniques that are obvious to those who understand how to create interactive web sites.
This is one example of a lack of quality control standards and proper understanding on the part of the USPTO regarding technology and the Internet. There are many more being awarded daily, as you can find out for yourself by doing a search for "patent" at http://www.slashdot.org/search.pl.
If elected, what are your plans, if any, for reform of the USPTO to prevent further abuse of our patent system? If you believe they are currently doing a good job and should not be reformed, how do you defend this position?
You can find many of these patent problems the subject of past slashdot discussions:
One Click Patent News
Barnes & Noble challenges Amazon 1-clic
Amazon takes round one in patent disupte
Matt Barnson
HOORAY!
Now, if only the Your Rights Online section had the readership of the rest of Slashdot, an article like this might deserve a fair shake.
Matt Barnson
First, about myself: I'm 27, I've been working mainly in the Internet industry since I was 21. College dropout because work was too interesting and my grades stunk. I was hired at my current company as a mid-level UNIX Systems Administrator, with two years UNIX experience (and five years Windows/DOS/Macintosh experience). I am now the manager of Systems Administration for Excite Business Applications in Utah.
I'll freely admit that I see age discrimination regularly. Many managers assume that older IT workers do not have the current skill sets to handle our work. I believe that's wrong, but having been in the business six years now I'm noticing that the average age is going up -- whereas most workers were under thirty before, I'm regularly seeing people into their forties and fifties in positions of responsbility at my company and elsewhere. I really feel that it's the *industry* which is finally growing up. There is an enormous shortage of competent, skilled workers in the field. From anecdotal evidence, I would suggest somewhere between five out of ten and nine out of ten applicants do not have any clue about their jobs. That's pretty sad, and makes us (managers) very jaded.
I'm looking for bright, competent workers in our entry-level systems administration positions. We hire more experienced admins as well; please allow me to offer a few tips on resumes that really make the difference for me on whether I ask someone to interview with me:
- Do not list time frames for previous positions on your resume. This avoids potential pre-interview age discrimination. Most employers aren't actively discriminating against older people, they just assume you're far too expensive for the position. Don't let them know your age on the resume.
- Remember, your resume is an advertisement about yourself, not a listing of your previous failures. Talk yourself up!
- List only relevant experience about yourself in your resume. If a job you held 15 years ago is relevant, list it. If the last job you held had nothing to do with the position for which you're interviewing, don't list it.
- The only way I'll believe your Summary of Qualifications is if I see work history that backs up each bullet point.
- Leverage your age as an asset in the interview. Convince the interviewer you are reasonably priced (their first concern when interviewing someone older than they are), more competent than any "young gun", and (most importantly!) able to learn quickly.
Many employers assume older workers are high priced, out-of date, and slower to learn than younger people. It's the prospective employee's job to recognize age discrimination exists; prove to the interviewer that these stereotypes do not apply to you. I hate to say it that way, but there are a lot of clueless newbie managers out there who simply don't know it's illegal, unethical, and stupid to discrimate based upon age.The law says all employment in the U.S. is "at will" -- you can leave, be terminated, or not get a job for any reason or no reason, outside of the following federally-protected exceptions:
I believe there are more, but cannot recall right now.
These are the things I, as an employer, willfully discriminate upon:
I wonder if I'm the exception rather than the rule? I certainly hope someone out there was benefitted by this post; if not, at the very least it helped me clarify my thoughts on this issue.
Matt Barnson
We have it, we like it. However, it costs $$, and has the potential of locking friends who value their privacy out of your life via phone...
Matt Barnson
The driver support under Linux for the WaveLAN card is fantastic, and the driver is even included with Linux-Mandrake 7.1 -- no special configuration required other than to make sure pcmcia inits before network.
The downside? It's not a 10-Mbps connection, it's only 2-Mbps. However, I routinely run X connections over SSH from a dedicated ssh gateway at my business. The link has fantastic uptime, the signal-noise ratio is almost as good as it gets with a 24dB antenna, and I'm completely undisturbed by power outages (UPS here), snow, rain, etc. High winds (60 m.p.h. + gusts outside right now) do take a toll, though. A 1 watt amplifier allows my 24dB antenna to punch right through the walls of my attic and lets me forget about the many trees, houses, and other obstructions between me and my ISP, 3 miles away.
The WaveLAN card with a pair of antennas and amps would be a much easier, more convenient option for most home users trying to set up this type of connection, IMHO. The amps are pretty expensive (a couple hundred dollars) but do wonders to ensure connection integrity.
Here's what my S/N ratio looks like, catting
A connection is destined for "www.excitestores.com", and ends up at the external DS/3 (T3, T1, insert your fast link here) port on our router. The router runs a rule against the packet and says "Hey, this is www traffic bound for the servers that are to be accelerated. Therefore my next hop is (insert IP address of cache here)!". It route-maps it to the cache server as it's next hop. The caching server is set up to "hijack" any incoming connections as if they are destined for itself, and makes the request to the origin web server on behalf of the requesting client. At this point, this does not differ too much from standard forward transparent proxying, except that you normally have an access control list that only permits transparent proxying of a limited set of URL's or IP addresses. You don't want to run an "open proxy" for the world to use to cache whatever they want.
Of course, note here that there are alternate methods of accelerating sites depending on the cache you choose and your infrastructure. The basic idea is to get the packets to your cache instead of the web server, however you choose to do it. Common methods include placing the cache in the natural route of the packets, making the webserver address point to the cache and have a non-public DNS that the cache looks to to resolve a web site on a non-routeable private network, or specifying on the cache that incoming connections on a certain IP are to accelerate a particular origin server.
Anyway, the benefits of this are enormous in our case. We have a (*&$load of modules compiled into our Apache server, tons of virtual hosts and modules to handle them all, and each daemon runs about 12 MB. Each web server has a gigabyte of RAM, therefore you do the math:
1024/12=85 and 1/3 connections run us out of physical RAM on each web server. Realize this is a rough estimate; our web servers can handle much more, but performance degrades quickly with more connections being served from virtual memory. I've also not taken into account OS overhead, other services running on the servers, and any other thing you may think of. However, modem users, particularly, saturate web server connections because it is so slow to deliver objects to them.
CNN.com, for instance, uses ICS caching boxes purely for connection management to handle these slower connections that could bog their servers down. Novell's ICS is rated at over 100,000 simultaneous connections on each box in reverse proxy mode. A big difference from 85 connections for one machine, no?
I'd love to discuss this in more depth, if you require a better answer. Better yet, check the FAQ at Squid's site regarding transparent reverse proxying.
Seriously, this is what takes web sites to the next level, regardless of whether you use Squid, ICS, NetCache, or another type of reverse cache. Keep smiling!
I've been overseeing a caching (really, website acceleration project) for my company, Excite@Home E-Business Services, over the last three months now. I can personally say that the three I've had experience with, Novell's ICS caches (which comprised ten of the twenty entrants), Network Appliance's NetCache, and Squid (on Solaris, in our case) all rock. Squid 2.3-stable1 was a dream to compile, install, and configure. ICS has a few user interface quirks with their Java administration tool that I don't like, but except for Cisco's cache (Oh My Gosh do you really want to spend $150,000 on a CACHE???) ICS-based systems captured the many top honors in this roundup. Network Appliance's NetCache is also a nice choice, and as the only vendor with streaming media caching/splitting support, they are receiving a lot of attention recently.
It's really important to note that IRCache has no desire to point to any "winner" in this bakeoff, but instead to have real non-partisan numbers to point to when evaluating cache performance. Squid captured top honors in cache hit ratios, but nothing else (AFAICT), showing that those "expensive, proprietary systems" also can be very well-tuned operating systems that eliminate traditional OS overhead for these numbers.
One of the frequently overlooked uses of cache is as a web site accelerator, instead of the standard forward proxy. Using a few simple access control lists and a policy on a router, reverse-proxy caches managed to reduce the instantaneous load on our web servers by up to 94%. We serve about 3.5 million hits a day. A "reverse proxy" is an EXCELLENT use of a proxy cache, and after these technology evaluations I've been involved with in past weeks I'd recommend it to anybody considering running a high-traffic website. This allows your Apache servers to function more as the "cgi engine" of your site, and lets the static images, text, banners, etc. be delivered from a box that can handle 100,000 simultaneous connections. Very cool.
While I'm not allowed to post a "review" of any one of these units, because of various agreements for the evaluation boxes we tested, I can clearly state that Squid, NetCache, and ICS-based systems can and will vastly reduce infrastructure scalability costs for businesses when deployed in a reverse-proxy configuration. Our earlier estimates guessed we'd need to expand our web farm three times to handle our estimated load by the end of the year. Now we can reliably predict that our farm can serve 10 times the amount of hits we're running now by using a cache as an accelerator. VERY cool stuff.
Be sure and check out the system configurations in the bakeoff review. It's very illustrative that the boxes tested have VERY specific audiences. Don't be fooled by the "fastest hit response time" or "most throughput" -- you can spend $6,000 or $150,000 for any setup, depending on your needs.
Noticeably absent from the review was Inktomi, for the second year in a row. I'm hearing FUD from vendors that their performance isn't up to snuff-- any truth to these rumors?
You know, we've run up against this same problem time and again in our new "connected" world. That the concepts of "taking" and of "property" lose meaning in the context of software. It appears our current judicial system just isn't well equipped to answer these questions.
As with many things, I think it depends on one's point of view.
From the author's point of view, if he views his algorithm as "property", it could be considered taking and he could demand payment. Alternately, since he's one who considers computer source code a form of speech (as do I), the restriction on exports of source code may be construed as a restriction on his first amendment rights.
It's a complex issue that isn't going to be solved in a day. The original issue with this guy was that the export restrictions, which were never voted upon nor approved by representatives of the people, but instead were handed down by Bill Clinton, a) should not apply to personal speech, and b) were an unconstitutional executive order.
I don't know what the answer is, but my representatives hear from me regularly on these kinds of issues. Remember that one consituent phone call is worth thousands of lobbyist dollars!
Sorry to ramble. Out.
I work for iMALL, Inc. which is soon to be acquired by Excite@Home. We just met today with Joe Kraus, one of the six co-founders of Excite. One of the questions asked was "Is there any truth to the rumor that Excite@Home is considering merging with AOL?"
The answer?
"No. I have no idea where this current rumor started. Well, I have some idea, but it really looks like they've recirculated old news about negotiations that were stopped three months before the rumor hit the papers."
He went on to say that yes, Excite@Home was talking with AOL months ago about putting portal links up on Excite pages for AOL/Netscape services. AOL thought Excite@Home's requirements were too high (a stock-sharing arrangement with AOL in exchange for the space) and the negotiations were dropped immediately (we're talking like 6 months ago).
The current restructuring is necessary to ease the burden on corporate heads and to bring divisions more inline with Excite@Home's underlying corporate philosophies and "the way things work" in the company. Things have been a bit uncomfortable since Excite's merger with @Home, but now the two complementary businesses (broadband access & search/portal business) support each other better, as well as make room for the upcoming merger with iMALL to serve as their e-commerce division.
I hope this has helped dispell this unfounded rumor.
Boy, one would think this guy knew his Marx and Lenin better...
AFAIK, Communism is rooted in the idea that the working man would rise up and overthrow the ruling class, distributing the fruits of labor equally to fellow proletariat. This man thinks the free software culture is communism?
Bah! Humbug!
The Internet is STILL a tiny club of culturally-elite, rich (compared to the rest of the world) burgoise representing everything despised by Marx. Despite our burgeoning population, we represent a tiny fraction of priveleged humanity trodding upon the backs of the repressed masses. We distribute our gifts to other members of this elite class of intellectuals.
The Free Software culture is far from Communism. It more resembles Nepotism, with the talented, rich few giving away to the less-talented rich few -- yet all members of the same, elite club. Maybe if we could bring the Internet to the huddled masses with no concept of computers...
But we can't even feed most of them.
I must agree with the comments noted in the review from others: The idea that the "gift culture" is communism is hogwash.
I've hand many chances to play with machines with redundant power supplies. In two words, THEY SUCK! Failure rate on redundant power supplies is ENORMOUS. In most cases (double entendre intended) two power supplies must occupy the space normally inhabited by a single power supply. They fail an order of magnitude more often than large single power supplies. IMHO, your overall best bet is redundant servers. Linux doesn't have many built-in high-availability features but the High-Availability HOWTO is a great place to start. But redundant power supplies -- PHEW! They stink!
We are deploying several hundred Pentium 2-class PCs throughout the district over the summer with the goal of every single classroom having an Internet-capable computer at their disposal. In addition, the District has just approved $3 million dollars for new hardware purchases so that each teacher may have a reasonable machine upon which to work and browse the Internet. Internet access is filtered (at least HTTP access) through the Utah Education Network to remove objectionable material. Internet access is provided at the head-end through a single T1. So, to answer your questions:
If you really, really want to help fix systems, not just mentor children in their use, in Salt Lake City area schools, contact the Granite School DistrictHuman Resources Department and let them know you wish to volunteer -- they'll definitely be able to help you out.
Good luck to you in your volunteering efforts.
I was one of the unwashed masses who were unfamiliar with Linux just a couple years ago. I was firmly in the Microsoft camp, doing daily drone drudgery to maintain PCs for customers. One day, while trying to find a cheap webserver, I discovered Linux and fell in love with it.
:), BeOS, or another OS, and Linux expertise is as common as donuts, I may consider it time to acquire some new "specialized skills" to make more money and be regarded as a "computer guru" with "the latest thing".
Does it feel good to be admired by your peers for competence with some "powerful, obscure" operating system? Absolutely! Is it fun to make comments in meetings such as "to do that would be trivial with Linux"? Of course! Does the fact that nobody else at work knows my chosen server and workstation operating system make me feel unique and special? Hell, yes!
From my experience:
*Knowing Linux and UNIX in general pays more than NT/NetWare expertise
*My knowledge of Linux makes me irreplaceable in the eyes co-workers and employers
*Linux does what I want it to do, the way I want to do it, for as long as I need it to, without crashing
When the Internet first went commercial, folks who knew HTML were paid big money. When people realized their applications weren't Y2K compliant, Cobol came back (briefly) into vogue. When the day comes that I get paid more for knowing Amiga
To sum up: I agree with the author, except that I think that there must, and will, always be enough new developers learning to work on Linux to supplant those who choose to move on. If not, Linux would become a footnote in the dusty tomes of history, with the UNIVAC, OS/2, and Charles Babbage's "Analytical Engine" (ca 1837).
In the right circumstances, a CD changer can be an important, vital storage device. Large CD Jukeboxes, like the types made at Meridian Data are perfect for long-term, near-line storage of critical data. Many models serve over 500 CD's (that's well over 300GB, quite an expensive set of hard drives) with anywhere from 2 to 8 CD-ROM drives allowing for multiple simultaneous access of different CDs. You can also get jukeboxes with CD-R drives for permanently burning important data to a media that has a shelf life of up to 50 years (better than an average hard drive's 10 years).
CD Changers and Jukeboxes are a vital part of any comprehensive Off-line/Near-line/On-line backup strategy, along with tape archives and hard drives. But laborious load times, hard user limits, and lamentable throughput will probably dog the media forever.
However, in the right situation it is the perfect solution.