Ok, they are running apache on windows I guess then? And that's the problem? Why are they running windows on "Unix Hardware"? What is "Unix hardware"? I can only assume they mean a Sun box? I didn't know Windows had a sparc version! I bet that's really awesome!
Anyway, from reading the article I get the impression that neither the interviewer nor the people interviewed have enough technical background to describe the problems accurately, much less fix them. The people interviewed are all managers who probably don't know the difference between c++ and VB, couldn't tell you what an OS actually is, or understand the difference between hardware and software (apparently).
In short, the story is that some managers who don't understand technology and were trying to deploy an apparently advanced web service for an entire school district never bothered to read the documentation of the software they were deploying, and then ran into trouble... I guess that's interesting, or news, or something..
Sometimes I'm happy that the ACLU et al are looking out for me, sometimes they pick the wrong fight. This is exactly one of them. Oh, packet 8 and vonage have 18 months to allow wire tapping? Guess what guys, they already have it. Vonage uses Silantro, its had calea support for at least the last 3 years. Broadworks (the Broadsoft softswitch) has calea as well. The large softswitch vendors all already support it, I think Asterisk even might (although I'm not sure). These things aren't going to make the "Internet more vulnerable to hackers".
Has the ACLU setup CALEA on these systems? I highly doubt it, but I have. At least with broadsoft it is a trivial matter to keep the softswitch entirely firewalled off the internet that unless someone finds a buffer overflow in the sip protocol or rtp protocol that the system is using there is no opportunity for a hacker to get in.
Furthermore, the system supporting CALEA doesn't increase the risk.. IE if someone hacks the SIP protocol stack on a softswitch and takes control of it, well who cares if the box supports CALEA they just got access to all the phonecalls going through that box.
Do you really thing that up til now the FBI et al has had no power to wire tap a VoIP phone? That more than 5 million people in the US are totally able to break whatever law they want (wire fraud, telemarketing scams, plan bank robberies, etc) notice I didn't mention terrorism, just because they have Vonage? Right.
I would say more than anything J2EE violates these patents. Basically it is a patent on the tiered architecture approach of putting a web server out front that hands requests off to an app server in the background. I've been doing setups like these with apache and tomcat/jboss for the last 5 years (although not before the patent was filed). I'd imagine someone was doing tiered architecture besides these guys in 98/97 etc?
Certainly ebay or amazon, altavista, one of the huge sites from the 90's has to have prior art on this. It's basically a load balancing setup.
It states in the article that Rackspace tried to turn over just the log files but then had to send the entire hard drive to comply with FBI rules.
Do any of you work for an ISP? I used to. If the FBI asks for logs like that you seriously have 12 hours generally to comply or the ISP is fined heavily. If they ask for something specific, and you're slogging through 6TBs of data, you can't possibly find exactly what the FBI wants in less than 12 hours.
The EFF lawyer says it would be like turning over a whole warehouse of documents instead of just one document... Well, good luck finding that needle in a haystack in 12 hours or face a fine that will bankrupt your company.
I don't know what business you're in that you have a pile of cash and you can just pay someone 100k/yr for 3 years to design and build something with no sales.. But I'm not in that business.
I started by myself, I ate because I got sales, if I didn't get sales I didn't pay rent (much less hire someone to design some new and unheard of thing). Now I've built my company up to 5 people total (myself and 4 employees). Right now, 95% of our revenue goes to expenses (payroll, benefits, etc), 5% is left over for future projects and growth, which amounts to a total of 20k/yr before taxes. I couldn't hire a receptionist for that, much less an engineer.
No lets just merge would not entail me keeping the executives seat. I dislike running a business. I enjoy the technology, I'm running a business today because I saw a need and decided to fill it, not because I'm some power hungry robber-baron.
The tiny tech outfits you speak of are all bankrupt. The ones who succeed are the MSs, Apples, and Googles of the world (oh yeah more of your hated corporations). And, the companies that sit around trying to "invent" something have either nice angel investors or VCs or a boatload of debt. Maybe in short, I don't know any company that sat there for "years" trying to make something and made it. Google got 100k the first time they showed someone their stuff, after spending exactly 0 dollars and a couple months nights and weekends hacking on it. MS got a deal to write Basic, Apple started with nothing but got going on their first working system. Amazon needed a ton of cash to get started, but that's what VCs and the stock market are for and still they bet the farm on 1 idea, they didn't sit around saying "Hey, we've got 4 billion dollars what should we do with it?"
Startups are cash strapped ideas, not idea poor cash hordes
Of course you've inferred entirely too much.. but oh well, my company employs 4 people (yeah not a million, but hey) all 4 of them are great people, who work hard and are paid fairly. I'm not some robber-baron, and I'm not suggesting that is the enlightened way. However, I know from running a small business, that somehow corraling 5, 10, 15, or 200 individual entities and making them work as one would be impossible without a central authority. You have to have a central authority, and by that I don't mean a dictatorship. Congress is a central authority, The judicial system (as a whole not just the supreme court) is a central authority. These are large complex bodies, but define a single thing (Congress the law, the courts the interpretation of the law). Just as Corporations aren't run by dictators boards of directors very often disagree, kick out CEOs, have falling outs among themselves and resign...
My problem with your idea isn't that its not dictatorial enough, its that as in all things moderation is the key. Fully decentralized small entities working together around some loosely defined goal, with no central authority to make final decisions, will not work. Further, if you elected 1 person from each company to go to a "Congress" you would never achieve success, because if the "Congress" is assigning tasks everyone is going to lobby everyone else for the most lucrative parts of the project, and in the end if people don't get what they want, they'll go work as a sub for NASA cause it pays better to design and build the next shuttle wing for nasa than to design the baggage storage system for xyz startup conglomerate.
I'm not against some great new idea of business. I would love it if there was a way I could compete more easily with the Ciscos, Avayas and Nortels of the world (if you couldn't guess I'm in the business phone system market). But, I don't think this is it. I wouldn't trust key parts of my business to some other entity without a rock solid contract between myself and them. And my dad's an attorney, I know how much rock solid contracts cost and I can't afford them. Neither can any of my competitors that are around my size. Cisco, Nortel, Avaya, those guys can afford rock solid contracts, so can Boeing... Small businesses can't.
not in all cases, however, if you have a network called linksys, or NETGEAR, or any of the other default ssids out there, that you regularly connect to, well guess what windows will happily connect to the strongest signal of any of those. At my office I setup new linksys ap's all the time, so my laptop always has linksys as one of the approved wireless nets. Around my parents house there are 4 (yes 4) neighbors who have linksys ap's all set to the defaults. My laptop always has fun there, it generally jumps between the 2 closest at least 3 or 4 times a day when I'm visiting... Anyway, point being in the most common setup (full defaults) windows XP sp2 will jump between like named networks depending on strength, and will automatically "hack" the nearest ap set with the same ssid as the one you normally use (if you're a defaulter yourself)
In my opinion we are discussing whether or not a large conglomerate of small companies can produce a large product (airplane, space shuttle, cancer drug) that requires a huge amount of upfront capital (5-10 billion) as economically and effectively as a single large corporation. I contend it can't be done without a powerful central authority the way a corp is run, if it can't be done as efficiently in a large group of small companies, then it is SOCIALLY IRRESPONSIBLE to do it because it wastes resources, and we get less safe planes that cost more and get here later. And I contend that because NO ONE has done it, it must therefore be less efficient, because if it were more efficient, people would be doing it and making money. Since they aren't, it must be less efficient, and therefore SOCIALLY IRRESPONSIBLE, and wasteful.
Granted, currently there are lots of examples to point at where large corps have failed, and if small businesses could do it better great more power to them. I just think the administrative overhead wasted on herding cats (the small businesses) would be too great, and in short, you'd have to basically turn everything into a state run project... you accuse me of loving dictatorships, I accuse you of loving communism.
My entire point is that just 2 or 3 small businesses teaming up isn't enough to change anything... in fact for small businesses merging is really the much more economical move. I own a small business, if I was approached by another small business who was interested in working with me in a quasi-partnership the way you are talking, I would say "no lets just merge" cause I don't have the budget or time to deal with an ongoing contract negotiation, and that's what your scheme calls for. Small businesses can't afford 100k/year attorneys to keep all the contracts up to date. We might be able to afford together an attorney for 1 month to do a merger, but not an ongoing partnership. Also, 2 or 3 small businesses don't have the capital to do anything big. You would have to have at least 150-200 companies the size of mine (about 400k/year in revenue) to be able to do anything remotely like design an airplane. And you can't start working on making an airplane until you have all 150-200 companies, and the capital to do it, because otherwise, all of your energy in a small business is focused on making enough money to pay the bills and make payroll. Small businesses don't have this huge reserve of cash just sitting there to spend on R&D. Economics 101 teaches you that (the main benefit of monopolies is that they have extra cash for R&D, and yes there are benefits to monopolies, they just generally don't outweight the negatives). In short your scheme is naive and neglects to account for the real world.
No all you need is a small group of companies (2 or 3) who begin the process by laying down plans and recruiting partners to the project and with them financing
I love the way you avoided only 1 argument in my last post. If this is "So easy" and only requires 2 or 3 companies getting together to start, and can produce any product or complete any project known or unknown to man, and is so much better at making money than large corps, and so much more efficient... then WHY, in A FREE MARKET, has NOBODY EVER DONE IT? If it really were so easy, and really would be so successful, it would have been tried by now and we'd all know the head guys, cause they'd all be billionaires. It would have dominated Boeing at aircraft, GM at cars, NASA at spaceflight, etc, etc...
Or did you seriously think that all the screws, washers, bolts, wires, connectors, lights, carpets, seat covers, tires, pipes, fluids etc etc etc on a 747 are made by Boing?
All of the things you mention above are provided by SUPPLIERS not sub-contractors. A Supplier has a vested interest in supplying parts on time, a sub-contractor does not (in fact subs often have a negative incentive to work quickly/be productive). There is a critical difference between a supplier and a sub-contractor. Suppliers are generally paid per part, subs are paid per hour, thus the more product a supplier provides, the more money they make... A sub on the other hand makes more money by taking longer to do the same job.
The 747 is designed, engineered, and tested by Boeing EMPLOYEES, it is manufactured and sold by Boeing EMPLOYEES. Yes they get parts from other companies, but it is not a sub-contractor/general contractor relationship, it is a supplier/purchaser relationship.
The only entities who rely heavily on sub-contractors are those in the infrastructure business (roads, buildings, homes, the government, and NASA). Those projects are always woefully over budget and late.
Further, in the car industry GM, Ford, etc are all vertically integrated. They all own almost all of their suppliers outright, generating more efficiencies, reducing the dependency on even the "suppliers" to get the job done right.
Yes, there are things that are inefficient about large corps, but if this "group of small businesses" you talk of could really compete with a corporation on projects such as this.. well, guess what? It is a free market, someone would have done it by now.
The one thing that you're not seeing is the HUGE and I do mean HUGE overhead of getting people from separate small companies to work peaceably together. How do you decide which company gets to R&D the wing and which gets the chair mount? Obviously the person doing the wing is going to make more money, so now you've got all your happy little companies in a bidding war against each other. How are you going to get 150 companies to listen to your "small administrative companies"? How do you make the company you told to R&D the cockpit work on that if they think they can do a better job on the tail section? And then you get into serious redundancy with 7 different wing designs submitted from 7 different companies, 4 different fusalages, 3 cockpits, 7 tails, 3 engine mounts.. because every company thinks its the best at doing xyz (not to mention xyz pays better than what the administrative company assigned them), and why are they gonna listen to some administrator from outside their company?
In the end you have a faction that breaks off and decides screw you guys we're gonna make our own plane... Well except they just took 150 million of your seed capital, and now noone has enough money to R&D a plane.
Either this idea is doomed by the above or its doomed by the fact that the only way to make people play nice is that everyone lawyers up, and everyone writes 1000 page contracts with everyone else (yeah sum(1...150) * 1000 pages... every company has to have a contract with every other company because there's no central authority.. in case you're wondering that's 11 million pages written by $400/hr attorneys.. that's alot of attorneys fees)
And even then, for the contracts to mean anything you've got to have the entire plan of action basically done already, you have to know up front which companies are going to be the best at wing design, which are the best at cockpit layout design, because which projects the companies are getting assigned have to be in the contracts. Thus you basically have to have a spontaneous event where suddenly 2000 people wake up one day and decide "OK we're gonna make a plane today, and last night I dreamed of all the other companies I have to partner with to make it happen".
All of the above goes away when you're all under one roof. There is a central authority who
The only problem with the above state approach is the huge amount of LAWYERS you just gave jobs. If you think you can run a cabal of 150 small businesses who are pooling resources to achieve something like hydrogen fuel cell vehicles with anything near the efficiency of a GM or Toyota or Honda.. Good freaking luck.
The contract negociations alone will cost 500 million easy just for the attorneys. Then you're gonna have a million contract disputes.. hell you'll spend GMs entire R&D budget for fuel cells over the next 5 years in the first year on attorneys alone, not to mention the 2000 employees of these small companies that are just sitting around waiting for the attorney's to say "ok go".
That is why corporations are necessary. Without the ability to pool and CONTROL an entire project centrally, large R&D, Infrastructure, etc (and by large I mean budgets in the 10s of billions) simply cannot be accomplished with anything getting close to efficiency. Just look at a simple thing like building a house. Because home builders rely entirely on subcontractors it is always over budget, behind schedule, and poorly done. And there are very very very often legal disputes.
The author also specifically mentions using these things in small setups. No IdM in a large enterprise is easy, simple or straight forward. However, if you've got a 50 seat network, I gaurantee you using samba+ldap on a linux server for IdM is much easier to manage, more straight forward to setup, and a whole lot cheaper than Windows Server 2003 Domain Controllers. Oh yeah, also it performs better.
Anyway, for large multi-site enterprises, nothing is simple (Novell, Windows, Linux, doesn't matter large is hard). For small businesses Active Directory is overkill (in price, features, and scalability). LDAP is a good fit there.
The only reason Europe and America can even compete at all in the global market place with China around is because the chinese gov't keeps its people opressed. If China were to become democratic, or its billion people could read, study, learn and do anything they wanted, it would take about 5 years before the chinese owned every major asset in the world and we'd all be their slaves.
I've been attempting to switch to Dvorak and I do notice a huge difference in comfort and even intuitivness when typing normal english like this post for example.
However I do alot of network admin work, commands like ping, and typing IP Addresses I find much more difficult in dvorak (I probably type between 300-400 IPs a day) because of the location of the period being so close to the number line. I also program alot and having the curly braces up on the number line is an extra stretch I don't have to make in qwerty. (although having the dash/underscore so close is nice). All in all for the work I do, I doubt dvorak saves me much in the RSI department.
The whole article basically sums up why patents don't work as intended. And I'm not talking about software patents, all patents. This field could have been huge 10 years ago, generating billions of dollars and furthering innovation. The supposed purpose of patents is to foster innovation and invention, alas, patents just stymie innovation for 20 years until they expire.
If as I've said before patents lasted 3 years, maybe 5 at the very most, they would probably be a good thing, in 3 years Iridian would have been able to establish itself as a market leader, and every newcomer to the field would most likely license their stuff anyway (under copyright, or some other license generated by the company). Instead it takes 20 years to get an iris scanner on my laptop, or built into a security system at my house? Those things should have been done in 92.
Ok, They replaced a stand-alone hardware firewall with ACLs in layer 3 switches.... Um, last time I set up a network with layer 3 ACLs it was significantly more time consuming and harder to manage than a firewall.
Further if they are using Virtualization technology and they mention each virtual server sits in a "DMZ" well, they they have to be running iptables or some other type of host based FIREWALL as there is not a physical network layer, the traffic isn't getting sent through a switch to do layer 3 ACL checking!
Next in this setup the client machines (yes often the most vulnerable and least looked after machines on any network) are left completely without network level protection, and rely on antivirus and vendor patches to be "safe". Maybe you can rely on MS to always release a patch before an exploit is released, I certainly don't.
Further, Antivirus technology is wholly flawed. What kind of protection does Norton offer you in the first 2 or 3 hours after a virus is released? It takes at least that long for them to analyze the threat and publish a new def for it, in the first 3 hours of any new large scale virus everyone on the planet who is relying on antivirus software is vulnerable.
These people are demonstrably retarded for setting their network up this way. This is certainly not the future of network security.
I agree with you, I'm not saying its not criminal to break into networks. I'm saying that if you want to be secure WEP+no dhcp+mac filtering is not enough.
If you want to conduct any sort of seriously private transactions over wireless, use a VPN to secure it, anything less just leaves your information out to be stolen.
hmm, encryption (WEP at least) broken in less than 10 seconds.
Password? Where do you add a password to a linksys wrt54g? Sure you can set an admin password but...
Mac spoofing is simple
once you crack the encryption its trivial to see broadcasts determine the network being used, and give yourself a static addy. I can and have gained access to a network protected as you state in less than 30 seconds.
The only way to protect yourself is to run a vpn over your wireless, and only quite nerdy people know how to set that up properly.
Further, once you have enabled all/any of these steps it becomes impossible for a friend to bring his laptop over and surf the net well not impossible, but who memorizes their 13 character WEP key, and their network subnet, and which addys are free, and wants to log in to their router everytime this happens and add a mac address?
At my house the above happens at least 3 times a week, I run a business out of my home and customers, co-workers, and contractors who work for me are constantly bringing new/different computers in to my home and need net access. Securing a WiFi network completely defeats accessibility/convienence factor that has made it so popular.
Now everyone is yelling at me saying I'm some sort of idiot, no, my "public" wifi is run out of a separate NIC on my firewall, it is completely untrusted, has a very small subnet (5 addys) handed out by DHCP, is fully logged, and allows traffic to pass only from the public net to the internet with dst port 80, it is purely for web browsing.
My actual home wifi network that I use is secured by openvpn, if you aren't connected to the VPN you aren't getting anywhere but my dummy page saying your access has been logged along with your mac address, please don't try that again.
I agree with you, I have 2TB at my home as well (well 1.75TB usable), 8 250GB SATA drives, at about 150 each, I actually think I have better price/GB. Anyway, having a huge disk array isn't even geeky its almost becoming a necessity (music, movies, tv shows, home videos), every home needs at least 500GB, probably closer to 1TB, and since drives are so big anymore, its easy to slap 4 or 5 in a case, get a SATA raid card, and fly.
Wow did you take MS's spin hook line and sinker. MS wants you to believe that's what all this genuine advantage crap is. Do you also believe it when MS tells you that windows 2000 is the most secure OS ever? That XP SP2 would solve all security problems?
I'm glad MS isn't just wasting all their marketing dollars apparently someone is listening. In reality all WGA is is a thinly vailed attempt by MS to extract more money from already paying customers. Do you think Dell, HP, Gateway, Toshiba, and Lenovo/IBM are distributing pirated copies of WinXP? What about Compusa, best buy and fry's? That is where probably 99% of MS's customers get their OS, and none of them are distributing pirated copies of windows. Sure your whitebox dealer on the corner might be, but most people don't buy their PCs there, and if they do, guess what, in good faith they thought they were getting a valid product, if it turns out they didn't, what do they have to do? Buy a whole new OS. It's not a deal for the consumer, its a deal for MS, and its a headache for anyone who changes hardware/software often.
I have one license for WinXP Pro, I move it regularly between 3 computers, and every time its another hour long session on hold and then getting them to reset the key so I can install it. It's totally legal, and the MS reps have always been helpful (If a bit slow getting to the phone), but its an hour (and $150) I lose every time I need to reinstall my OS. Sure I could buy 3 copies of the OS, but even then I change hardware so often that after the first month I'd have to be calling again on all 3 copies this time.
I jumped ship from the windows camp in 2000, and when I did I evaluated FreeBSD, OpenBSD, and 3 Linux distros to decide which one I would run on my home systems, which ones I would recommend to clients for solutions, etc.
I had 3 spare x86 boxes, and tinkered with all three OSs for at least 6 months.
I'm sorry Theo but the reason I use Linux instead of BSD has nothing to do with hating MS. I needed a better solution that was cheaper, I went looking, and Linux ran on 99% of my available hardware. FreeBSD while better than OpenBSD in the hardware support regard only supported about 60-70% of what I had at the time, and I could only get OpenBSD to even install on 1 of my spare machines, of course without sound or USB and I had to try 4 different NICs before I found a supported one.
Sure, the BSDs have better design, I agree, and I would love to run them, but if I'm limited to 10% of the available hardware and every time I need a new NIC I have to snoop around the store looking for that one magic NIC with the right chipset revision well I consider that a larger burden freedom-wise than MS places on its users. Stop yapping Theo and go write some firewire drivers, or whatever technology came out 5 years ago that your system still doesn't support.
Well, then they are somehow producing the G5 cheaper than dell produces its workstations, so kudos to them. Dell only has about 15% margins on workstations and 20% on servers, and a dual g5 is only about $50 more than a dual xeon workstation from dell, so for all of apple's small market they must really be raking those taiwanese manufacturers over the coals.
Hardware has alot of R&D too, or do you think apple just slaps a hard drive a mobo and a proc and some ram in a box and calls it done? Apple cannot hope to survive as a hardware company.
I think alot of you people are seriously confused about a little thing called profit margin.
Apple maybe clears 10% on hardware sales, software on the other hand commands 75-85% profit margins.
If they sell an iPod for $300, they might get to keep $30. If they sell OS X for $200, or iLife for $79, well OS X they'll get to keep $160, iLife they'll keep $60. So for every copy of iLife they sell, they have to sell 2 iPods just to make the same amount. Every copy of OS X they sell accounts for 5 iPods. Obviously the software market is a much better market to be in, and the only market it makes sense for Apple to try to grow.
Slashdot Mirror
Ok,
they are running apache on windows I guess then? And that's the problem? Why are they running windows on "Unix Hardware"? What is "Unix hardware"? I can only assume they mean a Sun box? I didn't know Windows had a sparc version! I bet that's really awesome!
Anyway, from reading the article I get the impression that neither the interviewer nor the people interviewed have enough technical background to describe the problems accurately, much less fix them. The people interviewed are all managers who probably don't know the difference between c++ and VB, couldn't tell you what an OS actually is, or understand the difference between hardware and software (apparently).
In short, the story is that some managers who don't understand technology and were trying to deploy an apparently advanced web service for an entire school district never bothered to read the documentation of the software they were deploying, and then ran into trouble... I guess that's interesting, or news, or something..
Yes there are. And besides all major softswitches already fully support CALEA (Silantro, Broadsoft, etc).
Sometimes I'm happy that the ACLU et al are looking out for me, sometimes they pick the wrong fight. This is exactly one of them. Oh, packet 8 and vonage have 18 months to allow wire tapping? Guess what guys, they already have it. Vonage uses Silantro, its had calea support for at least the last 3 years. Broadworks (the Broadsoft softswitch) has calea as well. The large softswitch vendors all already support it, I think Asterisk even might (although I'm not sure). These things aren't going to make the "Internet more vulnerable to hackers".
Has the ACLU setup CALEA on these systems? I highly doubt it, but I have. At least with broadsoft it is a trivial matter to keep the softswitch entirely firewalled off the internet that unless someone finds a buffer overflow in the sip protocol or rtp protocol that the system is using there is no opportunity for a hacker to get in.
Furthermore, the system supporting CALEA doesn't increase the risk.. IE if someone hacks the SIP protocol stack on a softswitch and takes control of it, well who cares if the box supports CALEA they just got access to all the phonecalls going through that box.
Do you really thing that up til now the FBI et al has had no power to wire tap a VoIP phone? That more than 5 million people in the US are totally able to break whatever law they want (wire fraud, telemarketing scams, plan bank robberies, etc) notice I didn't mention terrorism, just because they have Vonage? Right.
I would say more than anything J2EE violates these patents. Basically it is a patent on the tiered architecture approach of putting a web server out front that hands requests off to an app server in the background. I've been doing setups like these with apache and tomcat/jboss for the last 5 years (although not before the patent was filed). I'd imagine someone was doing tiered architecture besides these guys in 98/97 etc?
Certainly ebay or amazon, altavista, one of the huge sites from the 90's has to have prior art on this. It's basically a load balancing setup.
It states in the article that Rackspace tried to turn over just the log files but then had to send the entire hard drive to comply with FBI rules.
Do any of you work for an ISP? I used to. If the FBI asks for logs like that you seriously have 12 hours generally to comply or the ISP is fined heavily. If they ask for something specific, and you're slogging through 6TBs of data, you can't possibly find exactly what the FBI wants in less than 12 hours.
The EFF lawyer says it would be like turning over a whole warehouse of documents instead of just one document... Well, good luck finding that needle in a haystack in 12 hours or face a fine that will bankrupt your company.
I don't know what business you're in that you have a pile of cash and you can just pay someone 100k/yr for 3 years to design and build something with no sales.. But I'm not in that business.
I started by myself, I ate because I got sales, if I didn't get sales I didn't pay rent (much less hire someone to design some new and unheard of thing). Now I've built my company up to 5 people total (myself and 4 employees). Right now, 95% of our revenue goes to expenses (payroll, benefits, etc), 5% is left over for future projects and growth, which amounts to a total of 20k/yr before taxes. I couldn't hire a receptionist for that, much less an engineer.
No lets just merge would not entail me keeping the executives seat. I dislike running a business. I enjoy the technology, I'm running a business today because I saw a need and decided to fill it, not because I'm some power hungry robber-baron.
The tiny tech outfits you speak of are all bankrupt. The ones who succeed are the MSs, Apples, and Googles of the world (oh yeah more of your hated corporations). And, the companies that sit around trying to "invent" something have either nice angel investors or VCs or a boatload of debt. Maybe in short, I don't know any company that sat there for "years" trying to make something and made it. Google got 100k the first time they showed someone their stuff, after spending exactly 0 dollars and a couple months nights and weekends hacking on it. MS got a deal to write Basic, Apple started with nothing but got going on their first working system. Amazon needed a ton of cash to get started, but that's what VCs and the stock market are for and still they bet the farm on 1 idea, they didn't sit around saying "Hey, we've got 4 billion dollars what should we do with it?"
Startups are cash strapped ideas, not idea poor cash hordes
Of course you've inferred entirely too much.. but oh well, my company employs 4 people (yeah not a million, but hey) all 4 of them are great people, who work hard and are paid fairly. I'm not some robber-baron, and I'm not suggesting that is the enlightened way. However, I know from running a small business, that somehow corraling 5, 10, 15, or 200 individual entities and making them work as one would be impossible without a central authority. You have to have a central authority, and by that I don't mean a dictatorship. Congress is a central authority, The judicial system (as a whole not just the supreme court) is a central authority. These are large complex bodies, but define a single thing (Congress the law, the courts the interpretation of the law). Just as Corporations aren't run by dictators boards of directors very often disagree, kick out CEOs, have falling outs among themselves and resign...
My problem with your idea isn't that its not dictatorial enough, its that as in all things moderation is the key. Fully decentralized small entities working together around some loosely defined goal, with no central authority to make final decisions, will not work. Further, if you elected 1 person from each company to go to a "Congress" you would never achieve success, because if the "Congress" is assigning tasks everyone is going to lobby everyone else for the most lucrative parts of the project, and in the end if people don't get what they want, they'll go work as a sub for NASA cause it pays better to design and build the next shuttle wing for nasa than to design the baggage storage system for xyz startup conglomerate.
I'm not against some great new idea of business. I would love it if there was a way I could compete more easily with the Ciscos, Avayas and Nortels of the world (if you couldn't guess I'm in the business phone system market). But, I don't think this is it. I wouldn't trust key parts of my business to some other entity without a rock solid contract between myself and them. And my dad's an attorney, I know how much rock solid contracts cost and I can't afford them. Neither can any of my competitors that are around my size. Cisco, Nortel, Avaya, those guys can afford rock solid contracts, so can Boeing... Small businesses can't.
not in all cases, however, if you have a network called linksys, or NETGEAR, or any of the other default ssids out there, that you regularly connect to, well guess what windows will happily connect to the strongest signal of any of those. At my office I setup new linksys ap's all the time, so my laptop always has linksys as one of the approved wireless nets. Around my parents house there are 4 (yes 4) neighbors who have linksys ap's all set to the defaults. My laptop always has fun there, it generally jumps between the 2 closest at least 3 or 4 times a day when I'm visiting... Anyway, point being in the most common setup (full defaults) windows XP sp2 will jump between like named networks depending on strength, and will automatically "hack" the nearest ap set with the same ssid as the one you normally use (if you're a defaulter yourself)
In my opinion we are discussing whether or not a large conglomerate of small companies can produce a large product (airplane, space shuttle, cancer drug) that requires a huge amount of upfront capital (5-10 billion) as economically and effectively as a single large corporation. I contend it can't be done without a powerful central authority the way a corp is run, if it can't be done as efficiently in a large group of small companies, then it is SOCIALLY IRRESPONSIBLE to do it because it wastes resources, and we get less safe planes that cost more and get here later. And I contend that because NO ONE has done it, it must therefore be less efficient, because if it were more efficient, people would be doing it and making money. Since they aren't, it must be less efficient, and therefore SOCIALLY IRRESPONSIBLE, and wasteful.
Granted, currently there are lots of examples to point at where large corps have failed, and if small businesses could do it better great more power to them. I just think the administrative overhead wasted on herding cats (the small businesses) would be too great, and in short, you'd have to basically turn everything into a state run project... you accuse me of loving dictatorships, I accuse you of loving communism.
My entire point is that just 2 or 3 small businesses teaming up isn't enough to change anything... in fact for small businesses merging is really the much more economical move. I own a small business, if I was approached by another small business who was interested in working with me in a quasi-partnership the way you are talking, I would say "no lets just merge" cause I don't have the budget or time to deal with an ongoing contract negotiation, and that's what your scheme calls for. Small businesses can't afford 100k/year attorneys to keep all the contracts up to date. We might be able to afford together an attorney for 1 month to do a merger, but not an ongoing partnership. Also, 2 or 3 small businesses don't have the capital to do anything big. You would have to have at least 150-200 companies the size of mine (about 400k/year in revenue) to be able to do anything remotely like design an airplane. And you can't start working on making an airplane until you have all 150-200 companies, and the capital to do it, because otherwise, all of your energy in a small business is focused on making enough money to pay the bills and make payroll. Small businesses don't have this huge reserve of cash just sitting there to spend on R&D. Economics 101 teaches you that (the main benefit of monopolies is that they have extra cash for R&D, and yes there are benefits to monopolies, they just generally don't outweight the negatives). In short your scheme is naive and neglects to account for the real world.
No all you need is a small group of companies (2 or 3) who begin the process by laying down plans and recruiting partners to the project and with them financing
I love the way you avoided only 1 argument in my last post. If this is "So easy" and only requires 2 or 3 companies getting together to start, and can produce any product or complete any project known or unknown to man, and is so much better at making money than large corps, and so much more efficient... then WHY, in A FREE MARKET, has NOBODY EVER DONE IT? If it really were so easy, and really would be so successful, it would have been tried by now and we'd all know the head guys, cause they'd all be billionaires. It would have dominated Boeing at aircraft, GM at cars, NASA at spaceflight, etc, etc...
Or did you seriously think that all the screws, washers, bolts, wires, connectors, lights, carpets, seat covers, tires, pipes, fluids etc etc etc on a 747 are made by Boing?
All of the things you mention above are provided by SUPPLIERS not sub-contractors. A Supplier has a vested interest in supplying parts on time, a sub-contractor does not (in fact subs often have a negative incentive to work quickly/be productive). There is a critical difference between a supplier and a sub-contractor. Suppliers are generally paid per part, subs are paid per hour, thus the more product a supplier provides, the more money they make... A sub on the other hand makes more money by taking longer to do the same job.
The 747 is designed, engineered, and tested by Boeing EMPLOYEES, it is manufactured and sold by Boeing EMPLOYEES. Yes they get parts from other companies, but it is not a sub-contractor/general contractor relationship, it is a supplier/purchaser relationship.
The only entities who rely heavily on sub-contractors are those in the infrastructure business (roads, buildings, homes, the government, and NASA). Those projects are always woefully over budget and late.
Further, in the car industry GM, Ford, etc are all vertically integrated. They all own almost all of their suppliers outright, generating more efficiencies, reducing the dependency on even the "suppliers" to get the job done right.
Yes, there are things that are inefficient about large corps, but if this "group of small businesses" you talk of could really compete with a corporation on projects such as this.. well, guess what? It is a free market, someone would have done it by now.
The one thing that you're not seeing is the HUGE and I do mean HUGE overhead of getting people from separate small companies to work peaceably together. How do you decide which company gets to R&D the wing and which gets the chair mount? Obviously the person doing the wing is going to make more money, so now you've got all your happy little companies in a bidding war against each other. How are you going to get 150 companies to listen to your "small administrative companies"? How do you make the company you told to R&D the cockpit work on that if they think they can do a better job on the tail section? And then you get into serious redundancy with 7 different wing designs submitted from 7 different companies, 4 different fusalages, 3 cockpits, 7 tails, 3 engine mounts.. because every company thinks its the best at doing xyz (not to mention xyz pays better than what the administrative company assigned them), and why are they gonna listen to some administrator from outside their company?
In the end you have a faction that breaks off and decides screw you guys we're gonna make our own plane... Well except they just took 150 million of your seed capital, and now noone has enough money to R&D a plane.
Either this idea is doomed by the above or its doomed by the fact that the only way to make people play nice is that everyone lawyers up, and everyone writes 1000 page contracts with everyone else (yeah sum(1...150) * 1000 pages... every company has to have a contract with every other company because there's no central authority.. in case you're wondering that's 11 million pages written by $400/hr attorneys.. that's alot of attorneys fees)
And even then, for the contracts to mean anything you've got to have the entire plan of action basically done already, you have to know up front which companies are going to be the best at wing design, which are the best at cockpit layout design, because which projects the companies are getting assigned have to be in the contracts. Thus you basically have to have a spontaneous event where suddenly 2000 people wake up one day and decide "OK we're gonna make a plane today, and last night I dreamed of all the other companies I have to partner with to make it happen".
All of the above goes away when you're all under one roof. There is a central authority who
The only problem with the above state approach is the huge amount of LAWYERS you just gave jobs. If you think you can run a cabal of 150 small businesses who are pooling resources to achieve something like hydrogen fuel cell vehicles with anything near the efficiency of a GM or Toyota or Honda.. Good freaking luck.
The contract negociations alone will cost 500 million easy just for the attorneys. Then you're gonna have a million contract disputes.. hell you'll spend GMs entire R&D budget for fuel cells over the next 5 years in the first year on attorneys alone, not to mention the 2000 employees of these small companies that are just sitting around waiting for the attorney's to say "ok go".
That is why corporations are necessary. Without the ability to pool and CONTROL an entire project centrally, large R&D, Infrastructure, etc (and by large I mean budgets in the 10s of billions) simply cannot be accomplished with anything getting close to efficiency. Just look at a simple thing like building a house. Because home builders rely entirely on subcontractors it is always over budget, behind schedule, and poorly done. And there are very very very often legal disputes.
The author also specifically mentions using these things in small setups. No IdM in a large enterprise is easy, simple or straight forward. However, if you've got a 50 seat network, I gaurantee you using samba+ldap on a linux server for IdM is much easier to manage, more straight forward to setup, and a whole lot cheaper than Windows Server 2003 Domain Controllers. Oh yeah, also it performs better.
Anyway, for large multi-site enterprises, nothing is simple (Novell, Windows, Linux, doesn't matter large is hard). For small businesses Active Directory is overkill (in price, features, and scalability). LDAP is a good fit there.
The only reason Europe and America can even compete at all in the global market place with China around is because the chinese gov't keeps its people opressed. If China were to become democratic, or its billion people could read, study, learn and do anything they wanted, it would take about 5 years before the chinese owned every major asset in the world and we'd all be their slaves.
I've been attempting to switch to Dvorak and I do notice a huge difference in comfort and even intuitivness when typing normal english like this post for example.
However I do alot of network admin work, commands like ping, and typing IP Addresses I find much more difficult in dvorak (I probably type between 300-400 IPs a day) because of the location of the period being so close to the number line. I also program alot and having the curly braces up on the number line is an extra stretch I don't have to make in qwerty. (although having the dash/underscore so close is nice). All in all for the work I do, I doubt dvorak saves me much in the RSI department.
The whole article basically sums up why patents don't work as intended. And I'm not talking about software patents, all patents. This field could have been huge 10 years ago, generating billions of dollars and furthering innovation. The supposed purpose of patents is to foster innovation and invention, alas, patents just stymie innovation for 20 years until they expire.
If as I've said before patents lasted 3 years, maybe 5 at the very most, they would probably be a good thing, in 3 years Iridian would have been able to establish itself as a market leader, and every newcomer to the field would most likely license their stuff anyway (under copyright, or some other license generated by the company). Instead it takes 20 years to get an iris scanner on my laptop, or built into a security system at my house? Those things should have been done in 92.
Ok,
They replaced a stand-alone hardware firewall with ACLs in layer 3 switches.... Um, last time I set up a network with layer 3 ACLs it was significantly more time consuming and harder to manage than a firewall.
Further if they are using Virtualization technology and they mention each virtual server sits in a "DMZ" well, they they have to be running iptables or some other type of host based FIREWALL as there is not a physical network layer, the traffic isn't getting sent through a switch to do layer 3 ACL checking!
Next in this setup the client machines (yes often the most vulnerable and least looked after machines on any network) are left completely without network level protection, and rely on antivirus and vendor patches to be "safe". Maybe you can rely on MS to always release a patch before an exploit is released, I certainly don't.
Further, Antivirus technology is wholly flawed. What kind of protection does Norton offer you in the first 2 or 3 hours after a virus is released? It takes at least that long for them to analyze the threat and publish a new def for it, in the first 3 hours of any new large scale virus everyone on the planet who is relying on antivirus software is vulnerable.
These people are demonstrably retarded for setting their network up this way. This is certainly not the future of network security.
I agree with you, I'm not saying its not criminal to break into networks. I'm saying that if you want to be secure WEP+no dhcp+mac filtering is not enough.
If you want to conduct any sort of seriously private transactions over wireless, use a VPN to secure it, anything less just leaves your information out to be stolen.
hmm, encryption (WEP at least) broken in less than 10 seconds.
Password? Where do you add a password to a linksys wrt54g? Sure you can set an admin password but...
Mac spoofing is simple
once you crack the encryption its trivial to see broadcasts determine the network being used, and give yourself a static addy. I can and have gained access to a network protected as you state in less than 30 seconds.
The only way to protect yourself is to run a vpn over your wireless, and only quite nerdy people know how to set that up properly.
Further, once you have enabled all/any of these steps it becomes impossible for a friend to bring his laptop over and surf the net well not impossible, but who memorizes their 13 character WEP key, and their network subnet, and which addys are free, and wants to log in to their router everytime this happens and add a mac address?
At my house the above happens at least 3 times a week, I run a business out of my home and customers, co-workers, and contractors who work for me are constantly bringing new/different computers in to my home and need net access. Securing a WiFi network completely defeats accessibility/convienence factor that has made it so popular.
Now everyone is yelling at me saying I'm some sort of idiot, no, my "public" wifi is run out of a separate NIC on my firewall, it is completely untrusted, has a very small subnet (5 addys) handed out by DHCP, is fully logged, and allows traffic to pass only from the public net to the internet with dst port 80, it is purely for web browsing.
My actual home wifi network that I use is secured by openvpn, if you aren't connected to the VPN you aren't getting anywhere but my dummy page saying your access has been logged along with your mac address, please don't try that again.
I agree with you,
I have 2TB at my home as well (well 1.75TB usable), 8 250GB SATA drives, at about 150 each, I actually think I have better price/GB. Anyway, having a huge disk array isn't even geeky its almost becoming a necessity (music, movies, tv shows, home videos), every home needs at least 500GB, probably closer to 1TB, and since drives are so big anymore, its easy to slap 4 or 5 in a case, get a SATA raid card, and fly.
Wow did you take MS's spin hook line and sinker. MS wants you to believe that's what all this genuine advantage crap is. Do you also believe it when MS tells you that windows 2000 is the most secure OS ever? That XP SP2 would solve all security problems?
I'm glad MS isn't just wasting all their marketing dollars apparently someone is listening. In reality all WGA is is a thinly vailed attempt by MS to extract more money from already paying customers. Do you think Dell, HP, Gateway, Toshiba, and Lenovo/IBM are distributing pirated copies of WinXP? What about Compusa, best buy and fry's? That is where probably 99% of MS's customers get their OS, and none of them are distributing pirated copies of windows. Sure your whitebox dealer on the corner might be, but most people don't buy their PCs there, and if they do, guess what, in good faith they thought they were getting a valid product, if it turns out they didn't, what do they have to do? Buy a whole new OS. It's not a deal for the consumer, its a deal for MS, and its a headache for anyone who changes hardware/software often.
I have one license for WinXP Pro, I move it regularly between 3 computers, and every time its another hour long session on hold and then getting them to reset the key so I can install it. It's totally legal, and the MS reps have always been helpful (If a bit slow getting to the phone), but its an hour (and $150) I lose every time I need to reinstall my OS. Sure I could buy 3 copies of the OS, but even then I change hardware so often that after the first month I'd have to be calling again on all 3 copies this time.
I jumped ship from the windows camp in 2000, and when I did I evaluated FreeBSD, OpenBSD, and 3 Linux distros to decide which one I would run on my home systems, which ones I would recommend to clients for solutions, etc.
I had 3 spare x86 boxes, and tinkered with all three OSs for at least 6 months.
I'm sorry Theo but the reason I use Linux instead of BSD has nothing to do with hating MS. I needed a better solution that was cheaper, I went looking, and Linux ran on 99% of my available hardware. FreeBSD while better than OpenBSD in the hardware support regard only supported about 60-70% of what I had at the time, and I could only get OpenBSD to even install on 1 of my spare machines, of course without sound or USB and I had to try 4 different NICs before I found a supported one.
Sure, the BSDs have better design, I agree, and I would love to run them, but if I'm limited to 10% of the available hardware and every time I need a new NIC I have to snoop around the store looking for that one magic NIC with the right chipset revision well I consider that a larger burden freedom-wise than MS places on its users. Stop yapping Theo and go write some firewire drivers, or whatever technology came out 5 years ago that your system still doesn't support.
Well, then they are somehow producing the G5 cheaper than dell produces its workstations, so kudos to them. Dell only has about 15% margins on workstations and 20% on servers, and a dual g5 is only about $50 more than a dual xeon workstation from dell, so for all of apple's small market they must really be raking those taiwanese manufacturers over the coals.
Hardware has alot of R&D too, or do you think apple just slaps a hard drive a mobo and a proc and some ram in a box and calls it done? Apple cannot hope to survive as a hardware company.
I think alot of you people are seriously confused about a little thing called profit margin.
Apple maybe clears 10% on hardware sales, software on the other hand commands 75-85% profit margins.
If they sell an iPod for $300, they might get to keep $30. If they sell OS X for $200, or iLife for $79, well OS X they'll get to keep $160, iLife they'll keep $60. So for every copy of iLife they sell, they have to sell 2 iPods just to make the same amount. Every copy of OS X they sell accounts for 5 iPods. Obviously the software market is a much better market to be in, and the only market it makes sense for Apple to try to grow.