You could build a board equipped with hooks or partial holsters where you just slip phone and PDA into. You either use generously spaced (=open) hooks so you have to hold the board a bit upright to keep the stuff from falling out, or you could use a few straps (velcroed or buttoned down) to secure the gadgets from falling out of the holsters.
Working as Security Engineer and having worked as network admin myself I can see both sides: For a proper security audit you need to be able to view each end everything - for which you usually need "full admin" access. As network admin you do not want anyone to change things - especially not without the net crew knowing what is happening.
So when I do an audit I usually ask for a "supervised admin access". This is root access (usually a SUed shell) opened for me with a regular admin sitting behind my back, watching me. This way he can learn where to look and dig for security issues - and I can always ask about the whys and hows that usually crop up during an audit of a system.
On the other hand a company-internal security officer will want to be able to "stop things" because of security reasons. Simple solution: the security (wo)man gets the okay from company management to give you orders to shut down this or that service or system, or to disable users because of security problems.
This way security problems can be solved quickly (admins usually know the systems better than the security guy), and without further interfering with standard operations.
Why does he want to have an MP3 gadget to record the sounds? I'd recommend a plain old music cassette recorder ("Walkman"-type, e.g. http://www.prodcat.panasonic.com/shop/product.asp? sku=RQ-L30&CategoryID=223). You have 45 or 60 minutes recording for each side, media is EXTREMELY cheap (compared to any digital portable). Plus the recorder itself is cheap enough (~ $40) to be replaced easily in case is is dropped accidentally into the swamp. Some of these things even have voice-activated recording, so you can set it near the frogs, go away and just wait. Transfer to PC is done with any cheap soundcard.
According to the Transrapid info web page (German language) http://www.mvp.de/tr/eigen.html the system is ~6dB less noisy (1/4 perceived loudness) than a TGV or an ICE at same speed. At the same noise level the Transrapid can go ~100km/h faster than TGV or ICE.
The electric/magnetic field is comparable with a TV set in 3m distance - so not much. Depending on the Transrapid generation you either have high voltage lines within the rail (old) or static magnets for inductive energy transfer (new).
The trick with the latter: you only have a static local magnet field. The (fast) moving Transrapid percieves this local field as alternating magnetic field which is converted to electricity to power the carried batteries. No (dangerous) high voltage lines, no alternating magnetic fields.
The main competitor against the French TGV is the German ICE - both trains with conventional rail rechnology. The energy consumption of a Transrapid is less that the one of the (slower cruising) TGV.
Transrapid's main problm is the non-compatible track. If you decide against TGV or ICE after 10 years, you still can use the tracks for your slow trains whereas the Transrapid rail will be completely useless...
The Transrapid is not using superconducting coils - but conventional e-magnets instead (combined lift and propulsion system). If there is a power outage it will drop onto its carbon-fibre braking sleds after the internal, constantly re-charged batteries have run dry.
More info on the (German language, sorry) Transrapid project page:
http://www.mvp.de/
A few facts: 128m length for 562 passengers, it only takes 165 seconds (or 16km) to reach 430km/h due to its ~25MW power linear motor. The energy consumption and noise is measurably lower than a comparable train (German ICE, French TGV, Japanese Shinkansen).
The problem I see with FS problems (e.g. after brownout) is not the data loss itself. There I am with you. I have encountered FS that were unwriteable or even completely unusuable after a serious crash/problem. THIS is the problem I see.
So if you use simplistics (V)FAT you need an idiot-proof method to restore the FS (functionality) itself. While this could be done with a (possibly automatic) "format disk" function, I think users won't accept something like this on an appliance, even if they are used to problems like this on a PC.
Before you chosse the filesystem decide what you want.
If you opt for CD-ROM (iso9660) you do not have to care for brownouts and power-off "shutdown"s. You will not have to care for write routines - and have the plain audio-cd output as free bonus.
If you want a re-writeable MP3 player, you will have to take the following things into account:
Brownout - power failure during any phase of operation. This includes especially writing operations.
Small size/performance requirements - XFS may be a bit of overhead for the small CPU. What about the other FS?
Self-repair - if an(y) error occurs the user cannot interact due to a missing console. he FS will have to be corrected automagically without data loss!
So ext2 is a bad choice (missing journal/self-repair), FAT even worse (deadly self-crippling easily possible). ext3/ReiserFS might be an option, but I do not have personal experiences with that yet.
But Flash programs - at least the ones I encountered (e.g. the simulations for http://www.future-ceo.de/ if I remember correctly) were EXTREMELY slow. For example it was no problem for me to type (really used sentences and notes) faster than the computer was able to draw the characters. Heck, that machine (only 233MHz K6 but SCSI-monster) is fast enough for multi-track audio editing!
The last time I remember a "speed" like that, was on my rusty 4.77MHz 8086 (later upgraded to Nec V30) using MS Word 5.0 in (Hercules) graphics mode...
Do not send unsolicited mail - especially do not send "anonymous" mass mails. Let the customer feel that you care for him and his specific needs and interests.
So only send mail that is
Specifically customized to the interests of the targeted user (user, not only user group).
Is sent only to the users that explicitly allowed you to send mail to them.
Has a real (information) benefit for the customer.
Internet customers do not want to receive garbage. Plain, short and concize information is the key.
But how to get the "allowed" email addresses and customization information? Simple: provide a good website where users can access the information they are searching for easily. Give them the option to be updated on SPECIFIC news. E.g. when downloading patches/new drivers give them the OPTION (optional, not necessary input) to be notified when there are updates.
What does that give the user? He does not have to search for the current drivers - and still is not molested with garbage.
What's your benefit? Satisfied customer (good product service). Excellent customization info. Unmolested non-customers. And much cheaper.
As for user customization: do not ask much questions. Deduct from the point of interest. Someone downloading driver for the top-model graphics adapter usually is highly interested in graphics-intense gaming. So he may be interested in news on new (better) graphic adapters, supported games, etc.
Email users are highly selective - so do not flood them. Be short and do not mail too often.
Another good option is hosting FAQs and mailinglists for products or services (not necessarily only yours). With this people get to know and value you for your service. Make sure you keep a neutral stance there (do NOT use those for plain advertizing)!! But you may add a "signature" with your advertizings there - and sneak in a pointer to your appropriate product when AND ONLY WHEN asked for.
There are basically two applications that spring to my mind:
First, an "omnidirectional" wheel. Ok, for this you do not need 3D as 2D will be sufficient. Plus you need a pretty clean surface, else the motor suspension = drive axis will be clogged up with dirt. A smooth "wheel" (=ball) will be better for high motor efficiency, but will increase slipperiness, too. So I guess directly using this drive as motored wheel is not the best choice.
Second, yu can cinstruct a spherical joint similar to the human hip or shoulder joint. With this you limit the movement arc to a cone with ~140 degrees opening. Spherical joints are especially interesting for major static/suspending joints - like hip or shoulder. But there are the shortcomings of this design:
The major problem will be the low torque and missing self-locking. Self-locking means, that the system does not have to use energy to keep the joint in that position. Excellent example for this is the worm-gear: nearly completely self-locking and high torques possible. In comparison this 3D-"freely spinning" joint is a (low power) direct driven, low-torque, non-locking.
As you have to use distributed permanent (=low power) magnets, the torque cannot be increased much (compared to classical e-motors with exclusively electro-magnets). Plus - as you need a high number of e-magnets - the motor is quite heavy.
All in all a nice idea, but not a good choice for most current uses IMHO.
Or - as for Germany (according to a court decision) about 250 000 USD fine per incident.
The argumentation goes as follows: because it is so easy and cheap to do that, low fines will not be sufficient to stop an avalanche effect that will render the communication system (email) useless for the user.
In Germany you do not have to opt-OUT for electronic marketing (phone, fax, email). Instead the only legal way for companies is opt-IN for email lists. The only opt-OUT ("Robinson-Liste") is for classical bulk/snail mail. Here the (comparatively high) costs are automatically a limiting factor that is not given with email.
Wether to choose SMP or to cluster, all boils down to how you can parallelize your simulation. Heavy inter-nucleus dependencies do not rule out the one nor the other.
Three things are critical when parallelizing work:
Raw CPU power (per money) - here x86 are not too bad, neither are Alphas. Sparcs are less powerful here (but much better in bus throughput).
Throughput - how much data has to be moved to/from the node for each iteration.
Latency - how long do the single iterations run on a node - and how long does the node have to wait for answer from the controlling server.
SMP machines usually have better throughput and latency than raw CPU power - whereas clusters have more CPU power and weaker throughput and even worse latency.
The major advantage of cluster is that you can re-use them in multiple configurations (tree, (hyper)cube, pipeline, flat neighbourhood) - whichever meets your needs best.
As for cheap clustering: maybe you have enough clients/workstations that can be (mis)used as cluster during nighttime (and NICEd down during working hours) for a first test run?
All boils down on how (good) you can parallelize your simulation. If you cannot, multiple processors - in either configuration - won't help you.
As in the grease-and-oil mechanics: you can come an astonishing way with a minimal multi-purpose toolset. But specialized tools are simply much better to solve the job - or to solve it at all. The art is to know when to choose which tool.
Similar thing applies to programming languages: there is a reason why one usually chooses Fortran/C for number crunching, Assembler/C for system programming, Perl/PHP/CF for dynamic websites, etc.
So I recommend to learn as much *different* programming paradigms and languages as possible to know the better tools for each new job.
Okay, I confess I will be always carrying my trusty swiss army knife - for you maybe leatherman, SOG tool, or whatever multi-util-thingy you refer. But I know other tools and where they give me their specific advantage...
Allowing software and algorithms to be patentable in EU (which they are currently NOT) seems to give some means of defense to EU companies. Currently most patent struggles are cleared by mutual patent right exchange. The more patents you have, the more you can change - and thus defend yourself against even insane patents. But with algorithms and software not being patentable in the EU the EU(-only) companies do not own software patents to barter with.
This is the line of reasoning behind the organizations supporting the effort to allow suftware patents in the EU. But these people often fail to see that this will only bring more problems to them than advantages in the mid and long(er) run.
The first step - and one of the easier thing to do - will be to mount the system for hands-free access. For Palmpads a wrist/forearm strap-on is an obvious solution. With this you can look at the display (nearly) hands-free and operate the system one-handed. In "rough" usage do not forget to mount it inside a water- and dusttight (transparent) bag.
As Handspring's Visor already sports a (mostly unused) microphone, one could write an add-on that recognizes only a (very) few spoken commands (next, back, enter, menu) to navigate through documentation you load onto the Visor.
Cheap USB webcams (starting at $30) do surprisingly well even in "dark" surroundings. With tools like vidcat / w3cam you can produce a continuous stream of pics (my K6/233 + cheap-o-plast webcam ~ 1 pic/sec). You can either keep them or run some kind of surveillance software on it. The most primitive application is to use the PC/webcam combo as cheap slow-scan video recorder.
There are several tools existing for Win9x (Homewatch?) that start recording only if there is something changing on the video capture.
To prevent the thieves taking the evidence away along with the PC, some programs push the(se) images to a website. Alternatively you could throw a cat5 cable over to your neighbour and save the pics on his PC (and maybe vice versa).
Disadvantage: USB is good for only a few meters.
Beyond that a video grabber (BT848 based or similar) and el-cheapo CCD cameras (starting at ~$40 for b/w) can do the job.
With a simple relais-based box and some primitive software (hackin' LEDmeter) you should be able to switch between a number of cameras (up to 8 for zero extra electronics - up to 256 with some electronics on the parallel port).
So buy either an "Enterprise" Linux system - or one CD set per PC. And make sure you get a proper receipt for the Software (-CDs).
Or let a professional company issue you an official "Enterprise license" - in which they state that "you" have aquired an enterprise-wide license to use the software - licenses details provided on the installation media. Let them charge a nominal fee ($1 or so) for issuing this license (technically the cost for printing and shipping) - and make sure they provide a proper receipt.
With this you can prove the actual value of the software (and licenses). A proven value cannot be taxed by a tax officer - he will have to guess the value if no value is given. The latter is standard procedure in nearly any country.
I for example sometimes buy the currentmost (Debian/OpenBSD) CDs - for convenience and to deduct these costs OFF my tax (training material blabla).
First you talk about your client mix: Outlook, Netscape, Pine. As Pine is Unix-Only (if I am not mistaken), there seem to be Unix clients within your company. And Unix clients usually cannot connect to MAPI servers. So you will have to enable POP/IMAP connectors to the Exchange servers. Another source of problems and license fees.
Majordomo - I am nearly sure that Majordomo does not work on WinNT/Exchange. So you will have to migrate the automatic maillist operation done by Majordomo into an Exchange application. I personally do not know one, but I am dead sure that a migration (if possible at all) will not be a nice job.
Then there is new equipment. For Exchange you need WinNT/W2k plus enough licenses plus exchange plus licenses plus Outlook plus licenses. For HA you will need a HA server for NT - probably high-end Compaq servers with fibre-channel backbone and storage array - or something similar - at least for the central servers. The price is considerable.
Ah yes, Server load. In my old (consulting) company we had ~40 Exchange servers (Compaq 5500 and 7500 - all RAM/HD/CPU-maxed) for ~7000 people distributed accross the country. Now do a bit of calculation. The price is nice...
A carbon tube with a diameter of 4*10^-10m (= 0.4nm)
The Nobel Price 1996 for Chhemistry was awarded for the discovery of a soccer ball - a very small one though. The "Buckyball" only consists of 60 carbon atoms. Its synthesis and detection awarded Robert F. Curl, Harold W. Kroto und Richard E. Smalley the price.
(Download)
The atoms of this molecule are placed exactly like the sextangular leather patches of a soccer ball. With this C60 is another modification of carbon among graphite and diamond. Until now quite a number of carbon spheres of different sizes have been found. The smallest one of this Fullerenes called class of carbon molecules - the C20 - was just recently described in the magazine Nature (2000, 407, S.60-63.).
(picture)
A few years ago there was quite an euphoria about C60 and relatives, and its applications in various areas of science was lively discussed (e.g. superconduction and HIV proteasis inhibition). Even if today anyone can order C60 in chemistry wholesale shops ($500 per gramm) there never was a real technological breakthrough.
The latter is hoped to come with carbon nanotubes - molecules formed like tubes that consist exclusively of carbon (like C60). These tubes were discovered 1991 by the Japanese researcher Sumio Iijima. Nanotube research started as offspring of Fullerene chemistry and is on its best way to surpass it.
(picture)
Carbon nanotubes consist of concentric graphite shells - between one and fifty of them which gives a diameter of 1-50nm. But nanotubes can be up to a millimetre in length - and it is this anisotropy that is of importance for applications. Fibers made of nanotubes are said to have an enormous tensile strength (100times stronger than steel with 1/16th of steel's weight). The interior of these tubes can be filled with various material (e.g. hydrogen) - even chemical reactions are possible within these miniature test-tubes.
Because of teir special electronic properties these tubes could become an important part of future nano technology.
(picture)
In today's issue of Nature two research groups from Hongkong and Japan report independently about the(ir) world smallest nanotube (Nature, 2000, 408, S. 50 und S. 51). It only has a radius of 0.2 nm (nanometers, 2*10^-10m). According to theory scientists this radius is the ultimate as smaller tubes won't be stable any more.
The extreme curvature of the carbon shells creates unusual physical properties: along the axis these tubes a metal-like conductivity is postulated. Further properties of this smallest of all nanotubes have to be researched, but be prepared for suprprise.
While there are firewalls supporting this option (e.g. CheckPoint's Firewall-1) it is not considered to be a good idea as DNS is comparatively easy to compromise. Use IP addresses instead of DNS for the rulebase and real authentication mechanisms for authentication.
I guess you want to enable some kind of authentication so only a limited number of people can get outside? Then authenticate the people, not their hardware (PCs, identified by the MAC address). For a number of protocols (esp. HTTP and SMTP) there are good standard authentication models (HTTP basic authentication on the proxy for example) you can use on proxy or SMTP servers for this purpose. Then only allow the mail and proxy servers to go out through the firewall.
A second method will involve firewalls that support user-, session- or link-authentication for known and/or unknown protocols. The latter usually requires authenticating on a special telnet session or web site on the firewall itself: as long as the (telnet) session is open, the firewall allows packets from the same IP address through. Nearly all commercial firewalls support at least one of these authentication methods.
Basic question here: what do you want to protect/authenticate and how strong/circumventable shall the authentication mechanism be?
The debate about Echelon was always kept low-profile in Germany (and other non-UK Europe). Formal inquiry committees were (and still are) (to be) quite wishy-washy about the subject.
So this young MP (of the German "green" party) filed suit against unknown so there is more force and behind investigations. Plus investigations of a states attourney are not as easy to manipulate or quieten as the work of some political committee.
The lawsuit (as I understood) is not about patents (etc.) but about industry espionage. The confusion arose from the fact that the proper bill for industry espionage is just a part of the pantent & semiconductor bills...
Please do not post babelfish "translations". Most people know that crutch of a tool - the so-called translation often leads to misunderstandings as one can see from the follow-up postings.
Sure, with babelfish you are faster ("first post"-syndrome) but very messy. So please do wait for much more correct manual translations.
Slashdot Mirror
You could build a board equipped with hooks or partial holsters where you just slip phone and PDA into. You either use generously spaced (=open) hooks so you have to hold the board a bit upright to keep the stuff from falling out, or you could use a few straps (velcroed or buttoned down) to secure the gadgets from falling out of the holsters.
Working as Security Engineer and having worked as network admin myself I can see both sides: For a proper security audit you need to be able to view each end everything - for which you usually need "full admin" access. As network admin you do not want anyone to change things - especially not without the net crew knowing what is happening.
So when I do an audit I usually ask for a "supervised admin access". This is root access (usually a SUed shell) opened for me with a regular admin sitting behind my back, watching me. This way he can learn where to look and dig for security issues - and I can always ask about the whys and hows that usually crop up during an audit of a system.
On the other hand a company-internal security officer will want to be able to "stop things" because of security reasons. Simple solution: the security (wo)man gets the okay from company management to give you orders to shut down this or that service or system, or to disable users because of security problems.
This way security problems can be solved quickly (admins usually know the systems better than the security guy), and without further interfering with standard operations.
...has a ruggedized Organizer originating from the famous Series3 (I guess): water-, dust- and drop-proof - maybe a bit above your price range (http://www.enterprise.psion.com/public/products/w orkaboutmx.htm) but with complete com solution (http://www.enterprise.psion.com/public/products/v comm.htm)
The GSM phone Siemens M35i is advertized as splash-water resistant. But I am not sure wether it can be used within USA.
Why does he want to have an MP3 gadget to record the sounds? I'd recommend a plain old music cassette recorder ("Walkman"-type, e.g. http://www.prodcat.panasonic.com/shop/product.asp? sku=RQ-L30&CategoryID=223). You have 45 or 60 minutes recording for each side, media is EXTREMELY cheap (compared to any digital portable). Plus the recorder itself is cheap enough (~ $40) to be replaced easily in case is is dropped accidentally into the swamp. Some of these things even have voice-activated recording, so you can set it near the frogs, go away and just wait. Transfer to PC is done with any cheap soundcard.
The electric/magnetic field is comparable with a TV set in 3m distance - so not much. Depending on the Transrapid generation you either have high voltage lines within the rail (old) or static magnets for inductive energy transfer (new).
The trick with the latter: you only have a static local magnet field. The (fast) moving Transrapid percieves this local field as alternating magnetic field which is converted to electricity to power the carried batteries. No (dangerous) high voltage lines, no alternating magnetic fields.
The main competitor against the French TGV is the German ICE - both trains with conventional rail rechnology. The energy consumption of a Transrapid is less that the one of the (slower cruising) TGV.
Transrapid's main problm is the non-compatible track. If you decide against TGV or ICE after 10 years, you still can use the tracks for your slow trains whereas the Transrapid rail will be completely useless...
*bzzzt*
The Transrapid is not using superconducting coils - but conventional e-magnets instead (combined lift and propulsion system). If there is a power outage it will drop onto its carbon-fibre braking sleds after the internal, constantly re-charged batteries have run dry.
More info on the (German language, sorry) Transrapid project page:
http://www.mvp.de/
A few facts: 128m length for 562 passengers, it only takes 165 seconds (or 16km) to reach 430km/h due to its ~25MW power linear motor. The energy consumption and noise is measurably lower than a comparable train (German ICE, French TGV, Japanese Shinkansen).
The problem I see with FS problems (e.g. after brownout) is not the data loss itself. There I am with you. I have encountered FS that were unwriteable or even completely unusuable after a serious crash/problem. THIS is the problem I see.
So if you use simplistics (V)FAT you need an idiot-proof method to restore the FS (functionality) itself. While this could be done with a (possibly automatic) "format disk" function, I think users won't accept something like this on an appliance, even if they are used to problems like this on a PC.
If you opt for CD-ROM (iso9660) you do not have to care for brownouts and power-off "shutdown"s. You will not have to care for write routines - and have the plain audio-cd output as free bonus.
If you want a re-writeable MP3 player, you will have to take the following things into account:
So ext2 is a bad choice (missing journal/self-repair), FAT even worse (deadly self-crippling easily possible). ext3/ReiserFS might be an option, but I do not have personal experiences with that yet.
But Flash programs - at least the ones I encountered (e.g. the simulations for http://www.future-ceo.de/ if I remember correctly) were EXTREMELY slow. For example it was no problem for me to type (really used sentences and notes) faster than the computer was able to draw the characters. Heck, that machine (only 233MHz K6 but SCSI-monster) is fast enough for multi-track audio editing!
The last time I remember a "speed" like that, was on my rusty 4.77MHz 8086 (later upgraded to Nec V30) using MS Word 5.0 in (Hercules) graphics mode...
So only send mail that is
Internet customers do not want to receive garbage. Plain, short and concize information is the key.
But how to get the "allowed" email addresses and customization information? Simple: provide a good website where users can access the information they are searching for easily. Give them the option to be updated on SPECIFIC news. E.g. when downloading patches/new drivers give them the OPTION (optional, not necessary input) to be notified when there are updates.
What does that give the user? He does not have to search for the current drivers - and still is not molested with garbage.
What's your benefit? Satisfied customer (good product service). Excellent customization info. Unmolested non-customers. And much cheaper.
As for user customization: do not ask much questions. Deduct from the point of interest. Someone downloading driver for the top-model graphics adapter usually is highly interested in graphics-intense gaming. So he may be interested in news on new (better) graphic adapters, supported games, etc.
Email users are highly selective - so do not flood them. Be short and do not mail too often.
Another good option is hosting FAQs and mailinglists for products or services (not necessarily only yours). With this people get to know and value you for your service. Make sure you keep a neutral stance there (do NOT use those for plain advertizing)!! But you may add a "signature" with your advertizings there - and sneak in a pointer to your appropriate product when AND ONLY WHEN asked for.
There are basically two applications that spring to my mind:
First, an "omnidirectional" wheel. Ok, for this you do not need 3D as 2D will be sufficient. Plus you need a pretty clean surface, else the motor suspension = drive axis will be clogged up with dirt. A smooth "wheel" (=ball) will be better for high motor efficiency, but will increase slipperiness, too. So I guess directly using this drive as motored wheel is not the best choice.
Second, yu can cinstruct a spherical joint similar to the human hip or shoulder joint. With this you limit the movement arc to a cone with ~140 degrees opening. Spherical joints are especially interesting for major static/suspending joints - like hip or shoulder. But there are the shortcomings of this design:
The major problem will be the low torque and missing self-locking. Self-locking means, that the system does not have to use energy to keep the joint in that position. Excellent example for this is the worm-gear: nearly completely self-locking and high torques possible. In comparison this 3D-"freely spinning" joint is a (low power) direct driven, low-torque, non-locking.
As you have to use distributed permanent (=low power) magnets, the torque cannot be increased much (compared to classical e-motors with exclusively electro-magnets). Plus - as you need a high number of e-magnets - the motor is quite heavy.
All in all a nice idea, but not a good choice for most current uses IMHO.
Or - as for Germany (according to a court decision) about 250 000 USD fine per incident.
The argumentation goes as follows: because it is so easy and cheap to do that, low fines will not be sufficient to stop an avalanche effect that will render the communication system (email) useless for the user.
In Germany you do not have to opt-OUT for electronic marketing (phone, fax, email). Instead the only legal way for companies is opt-IN for email lists. The only opt-OUT ("Robinson-Liste") is for classical bulk/snail mail. Here the (comparatively high) costs are automatically a limiting factor that is not given with email.
Three things are critical when parallelizing work:
SMP machines usually have better throughput and latency than raw CPU power - whereas clusters have more CPU power and weaker throughput and even worse latency.
The major advantage of cluster is that you can re-use them in multiple configurations (tree, (hyper)cube, pipeline, flat neighbourhood) - whichever meets your needs best.
As for cheap clustering: maybe you have enough clients/workstations that can be (mis)used as cluster during nighttime (and NICEd down during working hours) for a first test run?
All boils down on how (good) you can parallelize your simulation. If you cannot, multiple processors - in either configuration - won't help you.
As in the grease-and-oil mechanics: you can come an astonishing way with a minimal multi-purpose toolset. But specialized tools are simply much better to solve the job - or to solve it at all. The art is to know when to choose which tool.
Similar thing applies to programming languages: there is a reason why one usually chooses Fortran/C for number crunching, Assembler/C for system programming, Perl/PHP/CF for dynamic websites, etc.
So I recommend to learn as much *different* programming paradigms and languages as possible to know the better tools for each new job.
Okay, I confess I will be always carrying my trusty swiss army knife - for you maybe leatherman, SOG tool, or whatever multi-util-thingy you refer. But I know other tools and where they give me their specific advantage...
Allowing software and algorithms to be patentable in EU (which they are currently NOT) seems to give some means of defense to EU companies. Currently most patent struggles are cleared by mutual patent right exchange. The more patents you have, the more you can change - and thus defend yourself against even insane patents. But with algorithms and software not being patentable in the EU the EU(-only) companies do not own software patents to barter with.
This is the line of reasoning behind the organizations supporting the effort to allow suftware patents in the EU. But these people often fail to see that this will only bring more problems to them than advantages in the mid and long(er) run.
The first step - and one of the easier thing to do - will be to mount the system for hands-free access. For Palmpads a wrist/forearm strap-on is an obvious solution. With this you can look at the display (nearly) hands-free and operate the system one-handed. In "rough" usage do not forget to mount it inside a water- and dusttight (transparent) bag.
As Handspring's Visor already sports a (mostly unused) microphone, one could write an add-on that recognizes only a (very) few spoken commands (next, back, enter, menu) to navigate through documentation you load onto the Visor.
Cheap USB webcams (starting at $30) do surprisingly well even in "dark" surroundings. With tools like vidcat / w3cam you can produce a continuous stream of pics (my K6/233 + cheap-o-plast webcam ~ 1 pic/sec). You can either keep them or run some kind of surveillance software on it. The most primitive application is to use the PC/webcam combo as cheap slow-scan video recorder.
There are several tools existing for Win9x (Homewatch?) that start recording only if there is something changing on the video capture.
To prevent the thieves taking the evidence away along with the PC, some programs push the(se) images to a website. Alternatively you could throw a cat5 cable over to your neighbour and save the pics on his PC (and maybe vice versa).
Disadvantage: USB is good for only a few meters.
Beyond that a video grabber (BT848 based or similar) and el-cheapo CCD cameras (starting at ~$40 for b/w) can do the job.
With a simple relais-based box and some primitive software (hackin' LEDmeter) you should be able to switch between a number of cameras (up to 8 for zero extra electronics - up to 256 with some electronics on the parallel port).
So buy either an "Enterprise" Linux system - or one CD set per PC. And make sure you get a proper receipt for the Software (-CDs).
Or let a professional company issue you an official "Enterprise license" - in which they state that "you" have aquired an enterprise-wide license to use the software - licenses details provided on the installation media. Let them charge a nominal fee ($1 or so) for issuing this license (technically the cost for printing and shipping) - and make sure they provide a proper receipt.
With this you can prove the actual value of the software (and licenses). A proven value cannot be taxed by a tax officer - he will have to guess the value if no value is given. The latter is standard procedure in nearly any country.
I for example sometimes buy the currentmost (Debian/OpenBSD) CDs - for convenience and to deduct these costs OFF my tax (training material blabla).
First you talk about your client mix: Outlook, Netscape, Pine. As Pine is Unix-Only (if I am not mistaken), there seem to be Unix clients within your company. And Unix clients usually cannot connect to MAPI servers. So you will have to enable POP/IMAP connectors to the Exchange servers. Another source of problems and license fees.
Majordomo - I am nearly sure that Majordomo does not work on WinNT/Exchange. So you will have to migrate the automatic maillist operation done by Majordomo into an Exchange application. I personally do not know one, but I am dead sure that a migration (if possible at all) will not be a nice job.
Then there is new equipment. For Exchange you need WinNT/W2k plus enough licenses plus exchange plus licenses plus Outlook plus licenses. For HA you will need a HA server for NT - probably high-end Compaq servers with fibre-channel backbone and storage array - or something similar - at least for the central servers. The price is considerable.
Ah yes, Server load. In my old (consulting) company we had ~40 Exchange servers (Compaq 5500 and 7500 - all RAM/HD/CPU-maxed) for ~7000 people distributed accross the country. Now do a bit of calculation. The price is nice...
Smaller is not possible
Michaela Simon 02.11.2000
A carbon tube with a diameter of 4*10^-10m (= 0.4nm)
The Nobel Price 1996 for Chhemistry was awarded for the discovery of a soccer ball - a very small one though. The "Buckyball" only consists of 60 carbon atoms. Its synthesis and detection awarded Robert F. Curl, Harold W. Kroto und Richard E. Smalley the price.
(Download)
The atoms of this molecule are placed exactly like the sextangular leather patches of a soccer ball. With this C60 is another modification of carbon among graphite and diamond. Until now quite a number of carbon spheres of different sizes have been found. The smallest one of this Fullerenes called class of carbon molecules - the C20 - was just recently described in the magazine Nature (2000, 407, S.60-63.).
(picture)
A few years ago there was quite an euphoria about C60 and relatives, and its applications in various areas of science was lively discussed (e.g. superconduction and HIV proteasis inhibition). Even if today anyone can order C60 in chemistry wholesale shops ($500 per gramm) there never was a real technological breakthrough.
The latter is hoped to come with carbon nanotubes - molecules formed like tubes that consist exclusively of carbon (like C60). These tubes were discovered 1991 by the Japanese researcher Sumio Iijima. Nanotube research started as offspring of Fullerene chemistry and is on its best way to surpass it.
(picture)
Carbon nanotubes consist of concentric graphite shells - between one and fifty of them which gives a diameter of 1-50nm. But nanotubes can be up to a millimetre in length - and it is this anisotropy that is of importance for applications. Fibers made of nanotubes are said to have an enormous tensile strength (100times stronger than steel with 1/16th of steel's weight). The interior of these tubes can be filled with various material (e.g. hydrogen) - even chemical reactions are possible within these miniature test-tubes.
Because of teir special electronic properties these tubes could become an important part of future nano technology.
(picture)
In today's issue of Nature two research groups from Hongkong and Japan report independently about the(ir) world smallest nanotube (Nature, 2000, 408, S. 50 und S. 51). It only has a radius of 0.2 nm (nanometers, 2*10^-10m). According to theory scientists this radius is the ultimate as smaller tubes won't be stable any more.
The extreme curvature of the carbon shells creates unusual physical properties: along the axis these tubes a metal-like conductivity is postulated. Further properties of this smallest of all nanotubes have to be researched, but be prepared for suprprise.
While there are firewalls supporting this option (e.g. CheckPoint's Firewall-1) it is not considered to be a good idea as DNS is comparatively easy to compromise. Use IP addresses instead of DNS for the rulebase and real authentication mechanisms for authentication.
I guess you want to enable some kind of authentication so only a limited number of people can get outside? Then authenticate the people, not their hardware (PCs, identified by the MAC address). For a number of protocols (esp. HTTP and SMTP) there are good standard authentication models (HTTP basic authentication on the proxy for example) you can use on proxy or SMTP servers for this purpose. Then only allow the mail and proxy servers to go out through the firewall.
A second method will involve firewalls that support user-, session- or link-authentication for known and/or unknown protocols. The latter usually requires authenticating on a special telnet session or web site on the firewall itself: as long as the (telnet) session is open, the firewall allows packets from the same IP address through. Nearly all commercial firewalls support at least one of these authentication methods.
Basic question here: what do you want to protect/authenticate and how strong/circumventable shall the authentication mechanism be?
The debate about Echelon was always kept low-profile in Germany (and other non-UK Europe). Formal inquiry committees were (and still are) (to be) quite wishy-washy about the subject.
So this young MP (of the German "green" party) filed suit against unknown so there is more force and behind investigations. Plus investigations of a states attourney are not as easy to manipulate or quieten as the work of some political committee.
The lawsuit (as I understood) is not about patents (etc.) but about industry espionage. The confusion arose from the fact that the proper bill for industry espionage is just a part of the pantent & semiconductor bills...
Please do not post babelfish "translations". Most people know that crutch of a tool - the so-called translation often leads to misunderstandings as one can see from the follow-up postings.
Sure, with babelfish you are faster ("first post"-syndrome) but very messy. So please do wait for much more correct manual translations.
Thanks
yabHuj