The responses I'm reading rather evenly fall on both sides for the same reasons!
* Get your Masters you will:
* be hired more readily - companies look for people with degrees.
* be paid more when you are hired for a net $ win
* Don't get your Masters, get a job you will:
* be hired more readily - companies assume people with Masters want to be paid more and screen them out.
* make 2 years of income which will more than offset the cost of grad school plus any increase in pay you might get
This isn't terribly surprising - any one person on/. has only worked for or been exposed to the hiring practices of a relatively small number of companies, even if they've been changing jobs for at the average tech rate (?) of 3 years or something.
The conclusion you should draw from this is:
* when you go to work, either way apply to lots of companies! Some might screen you out because you have a Masters or because you don't!
* salary probably has a lot more to do with your negotation skills, location, job market etc.
I have some other things for you to consider, given how open you left the question: "which is better?"
Obviously your professors think you should stay in school. After all - they all did, didn't they? It worked out for them. And they might have some non-obvious stake in you sticking around too.
Well, what do you mean by better? This is a pretty personal decision. Perhaps "make more money (immediately or in total)" isn't your deciding factor.
How much do you like your field? Many responses seemed to assume that you wanted to get a career as a sysadmin/network admin, which wouldn't match your degree (not that it matches any degrees really). Do you want to get a fair bit more depth in a more narrow subject area in your field of study? If you are unsure about your field, you should probably get out and work!
Consider that once you are working full time, it will be a LOT harder to go back to school even part time, even if you work for a company that pays for all of it. This will get even harder if you get married or have kids.
Maybe you are already in a serious relationship or have kids? That should drive you pretty solidly towards "get a job."
How much do you care about -where- you work? Do you want to work for a specific company or range of companies? Maybe do a straw poll of the company or companies you are interested in, and see if THEY prefer an advanced degree. In computer engineering you might want to go work for HP, Intel, AMD or IBM - in which case a PhD might not be out of the question. And you might even get paid accordingly.
How much did you pay for your undergrad? If you got a fancy degree from Caltech or MIT at $50k/yr you are going to have a hard time "upping" that with a Masters. On the other hand if you went to a "low end" school (this is by your field mind you! An affordable state school could have an excellent rep for your field) AND you have excellent grades it might be wise to try for an MIT Masters to top it off.
What is your funding situation? Would you be paying your way (or would someone else pay it for you?) or would you need to be a TA/RA? Obviously if you've got someone lined up to pay it for you that weighs pretty heavily in the "yes do it" side!
Interested in starting your own business? How does the degree factor into that? Timing - would a 2 year delay help or hurt? Funding - would the money you spend on your degree impact your ability to start the business? Or are you one of "those" people who find grad school the perfect environment in which to start a startup?
Some mentioned teaching - but I didn't see much mention of mention teaching college. I'm guessing since you didn't mention it that teaching high school is not under consideration or you'd be looking at the well-covered Education degree. Teaching college doesn't necessarily mean full time. Increasingly colleges are hiring adjunct facult
If you use computers, I know that you have run into software that totally sucks. In fact, not only did it totally suck, but every piece of software that came close to doing that thing sucks. Or you have run into wanting to do something that simply no software out there does.
There are still LOADS of gaps like this anywhere from tiny utility software up to enterprise level stuff. Pick one. Whatever one bugs you the most. Write some really good software. Open source it and sell support. Or don't.. whatever. Just write good software.
So you need some deadlines to keep you going? Not uncommon. Have someone do it for you (isn't that what you would do by contracting?). Either get yourself a partner (preferably someone who is keen on handling all the _other_ parts of creating and running a company in exchange for the possible rewards) who is also a good deadline-setter and will not let you slack. Or hire yourself a business coach if you do want to try your hand at the other aspects of running a company and just want someone to egg you on.
Read Paul Graham's essays for encouragement and why starting your own software company is (still) a good idea. http://www.paulgraham.com/
Oh yeah - ALSO find yourself another engaging hobby or two. They must involve at least the following:
Social interaction. Yes you need this. You cannot work in front of a computer at work and do programming all day and then come home and do it all night. Your boss made that rule for a _reason_ . In order for your creative programming side to flow the rest of your mind must be fed. If you just program all day every day for primary job and then your side job your productivity will drop like a rock. This should ideally involve more than one person - a significant other will severely cut into the time you can spend on the stuff you need (socializing with more than 1 person and getting outdoors (see below)). It is a trade off.
Get out. Out of the house. Out of buildings. Gardening maybe. Or hiking. Bicycling. Whatever appeals really. This is important for all the same reasons that social interaction is. It will tend to give your mind a break from thinking too heavily and the opportunity for creative thoughts to bubble up. It will also keep your body healthier. Not Olympic gymnast healthier. Heck - gardening will leave you a fat slob (if you are, and want to remain so), but it will bring your health up a slight notch nevertheless. If you want to be time-efficient, find a hobby that combines social activity plus getting out - this would possibly allow the space to date. But I do feel that doing something relatively mindless (BUT NOT IN FRONT OF A SCREEN - no video games and no TV. They are not mindless enough) is also fairly important even if it is only for a short amount of time..but regularly. At least once per week. Heck - just sit outside in a lawn chair in the sun and make chain mail. No thought involved, but you get fresh air and sun.
Remember, the hobby must be engaging enough that you will continue to do it in spite of the pull to spend all of your time in front of the computer. Try out a few and see which one sticks with you for a while. Plop a reminder in your calendar a few months down the line to start the programming part (ie: don't get so sucked into the hobby that it cuts off your original plans). Plop a reminder in your calendar a few months down the line to re-examine your hobby(ies).
Yes, this will severely cut back on the total amount of time that you spend in front of the computer programming. In fact, you might get only a tiny bit of code done per week (best done in extended-concentration burst I know - maybe one weeknight and 6-8 straight hours on one weekend day). But it will be much higher quality and you will get a LOT more done during that time.
If you are concerned about the time issues and you happen to watch TV cut it out. Watching TV fulfills neither of the requirements for a healthy body and mind needed for programming. If
..in other words, you want to hire a consulant. Any sysadmin worth their beans can do what they do from offsite. But there will always be some tasks involved in sysadminning that they need to be physically present for (or at very least have to walk someone else through it over the phone).
There may be some setup (upfront expense) involved in them setting themselves up for remote access.
There are many sysadmins out there consulting, full or part-time as you need. As you point out, trust is an issue. I would argue that continuity of service is probably also an issue. For both of those reasons I suggest a largish, well known and trusted consulting company that does primarily sysadminning. Preferably one that can place someone local to you with the right skillset so that they can occasionally come into the office as necessary.
The one I know is Collective Technologies. They have a large presence at the LISA conference (deinative sysadmin annual conference) and are well liked by the community. I also know a few people who work for them and enjoy working there and feel it is a good company-unusual in the "consulting company" world.
My advice to you if you are thinking about growing at all, to where you might hire a sysadmin full-time would be to hire an intern sysadmin in addition to the professional. Post and ad and get someone from a local college or community college for 10-20 hours/week, $15-20/hour on an hourly basis. They can be the person to work with the remote sysadmin on tasks that require hands-on. They can potentially be mentored by the remote sysadmin. They can see if sysadminning is for them (since there is no sysadmin major anywhere, you'll be hiring a CS, EE or MIS type person). If things work out in a few years, you can hire them full-time. If not, no great expense to you at all.
If you are saying "oh, but we sure don't need someone even for 10-20 hours per week, particularly if we also are going to be hiring an expensive consultant anyway!" Consider the intern as an investment in your future and consider that if you use technology at all, there is *always* work to do. If they aren't fixing things or putting out fires they can be writing scripts, tweaking or improving or thinking and designing your network strategically, and planning technology purchases and upgrades. Someone has got to do it, and in a small business the key players usually have much more immediately business critical things to put their minds on.
If your business is tech-oriented that is to say you have some geeks and at least one uber-geek. And your uber-geek is not already generating massive profit in some other way- ie: s/he has time on their hands then go with an Asterisk system. Your uber-geek would get a chance to have fun and learn and you would get a cheap (not counting the uber-geek's time) solution. Everybody wins.
Otherwise, go with a Talkswitch system.
I can definitely recommend the Talkswitch small office PBX system as a link in your overall VOIP system. I have deployed 2 of them and they are networkable up to 4 units with a total of 32 onsite extensions and 16 lines. The baseline unit is $1500 for 4 lines, 8 extensions.
The Talkswitch folks have definitely put some considerable thought into the small business market and what it needs - cheapish (at "business pricing") solutions that are easy to administer and expandable...right up until the point when your business is probably getting big enough to be looking to drop a lot more money on a phone system without any problems. They take analog incoming lines (currently) so you need to get your VOIP lines with "Analog Telephone Adaptors"... which is how most VOIP vendor sell their product anyway and the cost for the ATA is trivial. Then you plug in any analog telephone saving you the major premium of paying for VOIP handsets.
The administration for these units is done through a simple GUI and is trivial.
We currently using a mix of regular phone company phone lines and some VOIP lines.
Why not get an Asterisk turnkey system?
First off let me say I'm a huge supporter of open source and the business that I work for uses it everywhere that we can. So that is where my biases are.
From what I can see, the current offerings are more expensive than the talkswitch and while they over scads of extra features, I would argue that almost all of those features are utterly inconsequential to a small business. AND the turnkey asterisks do not have one feature that is important - plug and play incremental expansion. Meaning that you have to dump your eg: 16 extension Yonder system when you need 17 extensions (note: I haven't researched this totally).
The Asterisk turnkey systems I am seeing even with their easy to manage GUIs still seem a little edgy and hackerish and *most* small business don't have their own resident uber-geek to solve problems. The day *will* come when some Asterisk vendor sells the perfect replacement for the Talkswitch systems - one that will even supercede it by say offering the ability to grow incrementally to a much larger size, perhaps starting even smaller.. and at a cheaper price. But it is not here yet.
As for the actual question - what provider?
As for the actual question - I am really not sure it matters much. We are using Vonage. The big vendors are well known and you probably know them already.. AT&T, Packet8 etc. Stay away from the little vendors. Don't even go with a big vendor unless they offer a "business" service. Why? It is *NOT* worth the pocket change (for a business)/month that you will save by going with some unknown. It is not even worth the $10/month you would save by trying to use "residential" for your business service (unless of course your business is _really_ small as in being run out of your house with you and maybe one other employee). You are already pushing the envelope by using VOIP. In general it is worth the 10-30% overhead as a business to buy products and services designed for and marketed to businesses and to go with major brands. That doesn't mean you can't shop around, buy on sale etc, but if your business is so strapped you are seriously considering going with Rudolph Funkmeyer's $6/month VOIP service you should really be wondering if your business plan is working out and maybe you should just fold your hand and give up.
The market seems to be settling around $50/month for the first business VOIP line and $15 after that. With this approach you will save money over POTS for any kind of long distance usage or any more than 2 lines or both.
What I wonder is - the cost of Anti-Virus software included in all these "Linux vs Windows TCO" comparisons.
Microsoft themselves making AV software is tantamount to admitting that it is pretty much a requirement that you have AV software in order to run any Windows machine (I know I, and most other systems administrators wouldn't considering running Windows without it). At current market prices for Norton/McAfee, that adds about $40 for the first year (license plus 1 year virus signature updates) + $20/yr afterwards (for virus signature updates). Due to the mfr dropping support, you have to pay $40 every couple of years for a new version also. Admittedly you can get site licenses and buy licenses in bulk which reduces the cost.
I know I'm horribly ignorant about all of these things, but why not use HID bulbs for projectors? Everyone is so concerned about light output, heat, bulb life, etc. It seems like HID would be better on all of those counts.
Where I used to work we solved this exact problem using serial controlled power strips. These power strips had a number of outlets and a serial port, and you simply connected one computer to the serial port and using basic (scriptable) commands you could get it to power on and off any one or all of the power outlets.
Course you still have to worry about graceful shutdowns, but in this case these were machines with no remote access (at the time) - like Windows and Mac machines, so there was simply no way of cleanly shutting them down.
I *like* the USB master port. I think the terapin is ridiculously priced (particularly when compared to the Archos Recorder 20 - USB 2.0, smaller form factor, 20Gb,lighter and plays *and* records mp3s). The USB master port is a killer feature though - if it works as expected. In principle that should allow it to be directly connected to any slave device (eg: a digital camera) and data transferred. Seems like it would be extremely handy for backing up PDAs, and digital cameras.
Actually, it looks like device to device transfer will be built into some newer USB 2.0 devices. I would say slave to slave, but it looks like it is basically an integrated master/slave controller chipset that will be available: http://www.eetimes.com/story/OEG200111 06S0058
It will be called On-the-Go and it is specifically to work around this problem. Sounds great to me, I'd just love to have a Digital camera and be able to dump it to something like an Archos mp3/USB hdd.
My 5260 has a telnet interface as a previous poster described - you just set your interface (or a virtual interface if you use an OS capable of that) on your ethernet card to 10.0.0.2 and then telnet to 10.0.0.1.
The 5260 is a bridge (ethernet to err.. RADSL I think), so just because it is a bridge doens't mean it doesn't have that telnet interface.
Obviously if your OS is not capable of virtual interfaces so you can only have 1 IP per interface, then when you have your interface set to 10.0.0.2 you're not going to be able to get *past* the bridge, but that's okay, it is only temporary.
Also - the 5260 can be upgraded via firmware (it is out there somewhere) to be a router instead of a bridge. Neat, huh? Try www.dslreports.com to get a lot of info about these.
Try good old radio shack. I was just in there today, and until the 28th of July they have a sale on a nice unit with wheels, a sloped top, and a straight up/down bottom. I don't know how many units, but pretty small. It looked to be plenty sturdy (built for amplifiers)
Price was $59 on sale. They also had a bigger one which was still less than $100.
The drawback is that they only have 1 small one per store, if that. So if you don't get the in store one you have to order it from the catalog, and I imagine shipping isn't cheap.
Link saturation seems to be the point -- on the principle that - "I've already paid for all this bandwidth, so why can't I use it to the fullest extent. If not getting something right now, then getting something that I might be needing".
It actually seems like a neat idea to me, and one that I've thought of (although if it became commonplace, I'll be ISP's would be failing left and right) as being useful.
Another useful thing would be to be able to "queue" up downloads explicitly (instead of implicitly by reference) (I think you can with some napster clients, although not in the ideal way) so that when the link is non-busy it will start sucking down data.
The hardest part to implement I think would be to ensure that your link isn't fully saturated when you want to use it. Then you would see a delayed response as your clever "look-ahead" backed down -- would you really see a net positive speed gain then?
The algorithm would have to be something like:
Is there any traffic currently, or has there been in the last X minutes? If so, then only use
half or 3/4 of what is currently available (ie: left over from what is currently being used, if any is).
else if during non-standard usage hours
use 9/10ths of the link
else if during standard usage hours
use 3/4 of the link
Obviously there would be a CPU price to pay, but if you've already got a much more powerful firewall/proxy than you need, why not?
Can you get that book (or others) at your public library? Sure, you might have to wait a week or two, but it usually can be done. If you're a college student, frequently college libraries are linked on a national level (while public libraries are usually linked locally) and you can order books from almost anywhere.
I don't know if it would work, but I would certainly contribute.
The trouble is that the information is diverse and scattered -- how do you judge if a company is good or not? You could probably figure out a good rating system eg:
lowest - explicitly (stated) closed, and no drivers for alt OS
to highest - explicitly (stated) open, perhaps with patent, perhaps not, with open soucre developed w/ company time/money drivers.
Seems like it would be kinda worthless if it was all hearsay and the best thing would be to document each "point" eg: they *say* they are open source, here is a copy of the company mission statement. or: they produce freeBSD drivers, here is the link to the ftp site.
It would be great to be in a position to make not only compatability choices (x or y works well with Linux -- there are plenty of sites that will tell you that info) but choices based on the "moral/political" aspect of the company.
If it is designed as a site where users can add or remove points from a company by submitting documentation or links or something.. should work out pretty well.
I think "we" should spend a little less time posting our verbal spew on Slashdot and making LINUX more commercially ready. Or maybe not. Maybe "we" should spend our time how we want.
Maybe *you* are not the same as "we." You are lumping yourself in with this guy who did a port to dreamcast, and what have you yourself done besides this worthless post to Slashdot? Hopefully something. But if not, *I* don't care. Because *I* am spending my time posting a worthless followup to your waste of energy post. Because that is what *I* want to do.
And I should think that is exactly why this port to dreamcast was created. Because it was what he wanted to do.
So if *you* _think_ at all, and you _think_ that time should be put into making LINUX commercially viable (maybe, maybe not. That's not what Linus had in mind when he made it after all. Or maybe I don't know what Linus had in mind) then *you* will have to be the one to do it.
A fax connection is data anyway. Making a fax connection over an analog (voice) phone line was non-optimal in the first place. Why make it more complicated by doing data (fax) over voice (VoIP) over data (IP)? That's just plain ridiculous.
There are already fax over IP solutions available which work just fine. It would be stupid to do anything else.
Okay, but the announcement specifically talks about DVD playback. So if the hardware only accelerates MPEG then the software must take care of CSS. Presumably the closed-source portion of the software does this. So my question still remains -- will the software enforce zones or not?
It has been previously mentioned that many computer-based DVD solutions do not even enforce zones, it is just not publicitized, and of course a number of dedicated DVD players can be hacked so as not to enforce zones.
Does the software/hardware enforce zones, or is it zone-free? The press release makes no mention of this that I saw other than to say that the driver is released binary only for DVD security (probably required by their DVD licensing agreement) while the playback software will be open source (why couldn't they just make a plugin for xanim?).
And of course the other real question is was it a _Cuban_ cigar?
That may be the case, if so it is pretty neat and it is a very cheap device to have a custom drive in it. If that is the case though you should be able to easily do that simply using 2 drives (one reading while the other writes). That might actually end up being more expensive (particularly if you used larger drives, like I would), but that's not really the issue -- the issue is that is fun to mess around.:-)
I realize that Tivo is also partly selling a service which provides listings of all the shows and playtimes, etc which I would be more than happy to pay for independantly.
The real thing is being able to set up/mess around with your own hardware and software.
I'm a geek. This is great for a non-geek, and maybe many geeks would find it appealing as well. But as a geek I'm very happy to use or configure a general purpose computer to do all of this stuff. My most likely scenario would be one general-purpose computer _dedicated_ to all of the following, and one general-purpose computer for general use (web surfing, word processing, etc). The dedicated one would do: dvd tv/vcr (a-la Tivo if only the software were there!) internet radio (preferably using dsp rather than cpu) local mp3 storage file server(?)
As a geek I see no need to buy a dedicated device when I can set up an even better dedicated device myself (and have fun doing it).
The place where I see myself buying a dedicated device where my geekly skills aren't quite up to making my own is a portable device for the car/motorcycle that does (or has modules for) all of the following: gps w/uploadable mapping packet radio location system (why not!) mp3 (using a dsp, not cpu again) radar detector -- using dsp algorithms like the valentine one, maybe with multiple antennas, etc so that you can set up and mess with the software settings (hopefully even source code) yourself. weatherproof (for the motorcycle) low power (yay transmeta!)
All of these things in a small box that plugs into the car or motorcycle would be ideal.
It's inefficient to do any of these for more than one OS...
...if it is difficult to do so for more than one OS, otherwise why not? If Java were really there...
Maybe it is just a hole in the market for a "VGM" -- virtual gaming machine that is coded platform independant from the OpenGL/sound drivers up to a nice easy gaming API. Automatic updates can then happen on the frontside of the API, and the VGM would only need to be updated very infrequently.
I don't know if this comparison has been made before but it seems to me that the Open Source movement is much like the Homebrew Computer Club.
The ideals in the case of the Homebrew Computer Club (as described in the interview) were: "empowering people without the companies owning the computers" only in this case it is software. We are in a better situation today -- almost anyone can be "hardware empowered," but if you want to be software empowered -- forget it.
Keep in mind that when you buy a piece of software like Windows or Office you are actually just buying a "license to use" the software. The software itself is *owned* my Microsoft (or Large Company of your Choice). You don't own it any more than people in the 70's owned the large computers that they rented/begged time on. Closed source is just one way that this non-ownership and disempowerment is enforced.
Want to get around the ATM fee? It's easy but the catch is that you usually have to go into a bank (or some institution) to do it.
Your checking/money market account should come with an associated debit card. This uses the standard VISA network. Simply make take a cash advance on your "VISA" card. There is no charge for cash advances by any insitution that I am aware of, although there is usually a cap on the amount, it is also usually much higher than you could get from an ATM (or the same, $500), plus you can get it in the bills of your choice! What could be better?
I guess that using the VISA network only costs the end bank some tiny fee and they just don't care about it, and they figure your bank is going to ream you with charges for taking a cash advance (they usually do) on your credit card. There's no need to inform them that it is not a credit card -- as far as they are concerned, it is. In one place I told them it was a debit card and they didn't care. So.. have fun.
Slashdot Mirror
The responses I'm reading rather evenly fall on both sides for the same reasons!
* Get your Masters you will:
* be hired more readily - companies look for people with degrees.
* be paid more when you are hired for a net $ win
* Don't get your Masters, get a job you will:
* be hired more readily - companies assume people with Masters want to be paid more and screen them out.
* make 2 years of income which will more than offset the cost of grad school plus any increase in pay you might get
This isn't terribly surprising - any one person on /. has only worked for or been exposed to the hiring practices of a relatively small number of companies, even if they've been changing jobs for at the average tech rate (?) of 3 years or something.
The conclusion you should draw from this is:
* when you go to work, either way apply to lots of companies! Some might screen you out because you have a Masters or because you don't!
* salary probably has a lot more to do with your negotation skills, location, job market etc.
I have some other things for you to consider, given how open you left the question: "which is better?"
Obviously your professors think you should stay in school. After all - they all did, didn't they? It worked out for them. And they might have some non-obvious stake in you sticking around too.
Well, what do you mean by better? This is a pretty personal decision. Perhaps "make more money (immediately or in total)" isn't your deciding factor.
How much do you like your field? Many responses seemed to assume that you wanted to get a career as a sysadmin/network admin, which wouldn't match your degree (not that it matches any degrees really). Do you want to get a fair bit more depth in a more narrow subject area in your field of study? If you are unsure about your field, you should probably get out and work!
Consider that once you are working full time, it will be a LOT harder to go back to school even part time, even if you work for a company that pays for all of it. This will get even harder if you get married or have kids.
Maybe you are already in a serious relationship or have kids? That should drive you pretty solidly towards "get a job."
How much do you care about -where- you work? Do you want to work for a specific company or range of companies? Maybe do a straw poll of the company or companies you are interested in, and see if THEY prefer an advanced degree. In computer engineering you might want to go work for HP, Intel, AMD or IBM - in which case a PhD might not be out of the question. And you might even get paid accordingly.
How much did you pay for your undergrad? If you got a fancy degree from Caltech or MIT at $50k/yr you are going to have a hard time "upping" that with a Masters. On the other hand if you went to a "low end" school (this is by your field mind you! An affordable state school could have an excellent rep for your field) AND you have excellent grades it might be wise to try for an MIT Masters to top it off.
What is your funding situation? Would you be paying your way (or would someone else pay it for you?) or would you need to be a TA/RA? Obviously if you've got someone lined up to pay it for you that weighs pretty heavily in the "yes do it" side!
Interested in starting your own business? How does the degree factor into that? Timing - would a 2 year delay help or hurt? Funding - would the money you spend on your degree impact your ability to start the business? Or are you one of "those" people who find grad school the perfect environment in which to start a startup?
Some mentioned teaching - but I didn't see much mention of mention teaching college. I'm guessing since you didn't mention it that teaching high school is not under consideration or you'd be looking at the well-covered Education degree. Teaching college doesn't necessarily mean full time. Increasingly colleges are hiring adjunct facult
If you use computers, I know that you have run into software that totally sucks. In fact, not only did it totally suck, but every piece of software that came close to doing that thing sucks. Or you have run into wanting to do something that simply no software out there does.
There are still LOADS of gaps like this anywhere from tiny utility software up to enterprise level stuff. Pick one. Whatever one bugs you the most. Write some really good software. Open source it and sell support. Or don't.. whatever. Just write good software.
So you need some deadlines to keep you going? Not uncommon. Have someone do it for you (isn't that what you would do by contracting?). Either get yourself a partner (preferably someone who is keen on handling all the _other_ parts of creating and running a company in exchange for the possible rewards) who is also a good deadline-setter and will not let you slack. Or hire yourself a business coach if you do want to try your hand at the other aspects of running a company and just want someone to egg you on.
Read Paul Graham's essays for encouragement and why starting your own software company is (still) a good idea. http://www.paulgraham.com/
Oh yeah - ALSO find yourself another engaging hobby or two. They must involve at least the following:
Social interaction. Yes you need this. You cannot work in front of a computer at work and do programming all day and then come home and do it all night. Your boss made that rule for a _reason_ . In order for your creative programming side to flow the rest of your mind must be fed. If you just program all day every day for primary job and then your side job your productivity will drop like a rock. This should ideally involve more than one person - a significant other will severely cut into the time you can spend on the stuff you need (socializing with more than 1 person and getting outdoors (see below)). It is a trade off.
Get out. Out of the house. Out of buildings. Gardening maybe. Or hiking. Bicycling. Whatever appeals really. This is important for all the same reasons that social interaction is. It will tend to give your mind a break from thinking too heavily and the opportunity for creative thoughts to bubble up. It will also keep your body healthier. Not Olympic gymnast healthier. Heck - gardening will leave you a fat slob (if you are, and want to remain so), but it will bring your health up a slight notch nevertheless. If you want to be time-efficient, find a hobby that combines social activity plus getting out - this would possibly allow the space to date. But I do feel that doing something relatively mindless (BUT NOT IN FRONT OF A SCREEN - no video games and no TV. They are not mindless enough) is also fairly important even if it is only for a short amount of time..but regularly. At least once per week. Heck - just sit outside in a lawn chair in the sun and make chain mail. No thought involved, but you get fresh air and sun.
Remember, the hobby must be engaging enough that you will continue to do it in spite of the pull to spend all of your time in front of the computer. Try out a few and see which one sticks with you for a while. Plop a reminder in your calendar a few months down the line to start the programming part (ie: don't get so sucked into the hobby that it cuts off your original plans). Plop a reminder in your calendar a few months down the line to re-examine your hobby(ies).
Yes, this will severely cut back on the total amount of time that you spend in front of the computer programming. In fact, you might get only a tiny bit of code done per week (best done in extended-concentration burst I know - maybe one weeknight and 6-8 straight hours on one weekend day). But it will be much higher quality and you will get a LOT more done during that time.
If you are concerned about the time issues and you happen to watch TV cut it out. Watching TV fulfills neither of the requirements for a healthy body and mind needed for programming. If
..in other words, you want to hire a consulant. Any
sysadmin worth their beans can do what they do from offsite. But there will always be some tasks involved in sysadminning that they need to be physically present for (or at very least have to walk someone else through it over the phone).
There may be some setup (upfront expense) involved in them setting themselves up for remote access.
There are many sysadmins out there consulting, full or part-time as you need. As you point out, trust is an issue. I would argue that continuity of service is probably also an issue. For both of those reasons I suggest a largish, well known and trusted consulting company that does primarily sysadminning. Preferably one that can place someone local to you with the right skillset so that they can occasionally come into the office as necessary.
The one I know is Collective Technologies. They have a large presence at the LISA conference (deinative sysadmin annual conference) and are well liked by the community. I also know a few people who work for them and enjoy working there and feel it is a good company-unusual in the "consulting company" world.
My advice to you if you are thinking about growing at all, to where you might hire a sysadmin full-time would be to hire an intern sysadmin in addition to the professional. Post and ad and get someone from a local college or community college for 10-20 hours/week, $15-20/hour on an hourly basis. They can be the person to work with the remote sysadmin on tasks that require hands-on. They can potentially be mentored by the remote sysadmin. They can see if sysadminning is for them (since there is no sysadmin major anywhere, you'll be hiring a CS, EE or MIS type person). If things work out in a few years, you can hire them full-time. If not, no great expense to you at all.
If you are saying "oh, but we sure don't need someone even for 10-20 hours per week, particularly if we also are going to be hiring an expensive consultant anyway!" Consider the intern as an investment in your future and consider that if you use technology at all, there is *always* work to do. If they aren't fixing things or putting out fires they can be writing scripts, tweaking or improving or thinking and designing your network strategically, and planning technology purchases and upgrades. Someone has got to do it, and in a small business the key players usually have much more immediately business critical things to put their minds on.
If your business is tech-oriented that is to say you have some geeks and at least one uber-geek. And your uber-geek is not already generating massive profit in some other way- ie: s/he has time on their hands then go with an Asterisk system. Your uber-geek would get a chance to have fun and learn and you would get a cheap (not counting the uber-geek's time) solution. Everybody wins.
Otherwise, go with a Talkswitch system.
I can definitely recommend the Talkswitch small office PBX system as a link in your overall VOIP system. I have deployed 2 of them and they are networkable up to 4 units with a total of 32 onsite extensions and 16 lines. The baseline unit is $1500 for 4 lines, 8 extensions.
http://talkswitch.com/
The Talkswitch folks have definitely put some considerable thought into the small business market and what it needs - cheapish (at "business pricing") solutions that are easy to administer and expandable...right up until the point when your business is probably getting big enough to be looking to drop a lot more money on a phone system without any problems.
They take analog incoming lines (currently) so you need to get your VOIP lines with "Analog Telephone Adaptors"... which is how most VOIP vendor sell their product anyway and the cost for the ATA is trivial. Then you plug in any analog telephone saving you the major premium of paying for VOIP handsets.
The administration for these units is done through a simple GUI and is trivial.
We currently using a mix of regular phone company phone lines and some VOIP lines.
Why not get an Asterisk turnkey system?
First off let me say I'm a huge supporter of open source and the business that I work for uses it everywhere that we can. So that is where my biases are.
From what I can see, the current offerings are more expensive than the talkswitch and while they over scads of extra features, I would argue that almost all of those features are utterly inconsequential to a small business. AND the turnkey asterisks do not have one feature that is important - plug and play incremental expansion. Meaning that you have to dump your eg: 16 extension Yonder system when you need 17 extensions (note: I haven't researched this totally).
The Asterisk turnkey systems I am seeing even with their easy to manage GUIs still seem a little edgy and hackerish and *most* small business don't have their own resident uber-geek to solve problems. The day *will* come when some Asterisk vendor sells the perfect replacement for the Talkswitch systems - one that will even supercede it by say offering the ability to grow incrementally to a much larger size, perhaps starting even smaller.. and at a cheaper price. But it is not here yet.
As for the actual question - what provider?
As for the actual question - I am really not sure it matters much. We are using Vonage. The big vendors are well known and you probably know them already.. AT&T, Packet8 etc. Stay away from the little vendors. Don't even go with a big vendor unless they offer a "business" service. Why? It is *NOT* worth the pocket change (for a business)/month that you will save by going with some unknown. It is not even worth the $10/month you would save by trying to use "residential" for your business service (unless of course your business is _really_ small as in being run out of your house with you and maybe one other employee). You are already pushing the envelope by using VOIP. In general it is worth the 10-30% overhead as a business to buy products and services designed for and marketed to businesses and to go with major brands. That doesn't mean you can't shop around, buy on sale etc, but if your business is so strapped you are seriously considering going with
Rudolph Funkmeyer's $6/month VOIP service you should really be wondering if your business plan is working out and maybe you should just fold your hand and give up.
The market seems to be settling around $50/month for the first business VOIP line and $15 after that. With this approach you will save money over POTS for any kind of long distance usage or any more than 2 lines or both.
What I wonder is - the cost of Anti-Virus software included in all these "Linux vs Windows TCO" comparisons.
Microsoft themselves making AV software is tantamount to admitting that it is pretty much a requirement that you have AV software in order to run any Windows machine (I know I, and most other systems administrators wouldn't considering running Windows without it). At current market prices for Norton/McAfee, that adds about $40 for the first year (license plus 1 year virus signature updates) + $20/yr afterwards (for virus signature updates). Due to the mfr dropping support, you have to pay $40 every couple of years for a new version also. Admittedly you can get site licenses and buy licenses in bulk which reduces the cost.
I know I'm horribly ignorant about all of these things, but why not use HID bulbs for projectors? Everyone is so concerned about light output, heat, bulb life, etc. It seems like HID would be better on all of those counts.
Where I used to work we solved this exact problem using serial controlled power strips. These power strips had a number of outlets and a serial port, and you simply connected one computer to the serial port and using basic (scriptable) commands you could get it to power on and off any one or all of the power outlets.
Course you still have to worry about graceful shutdowns, but in this case these were machines with no remote access (at the time) - like Windows and Mac machines, so there was simply no way of cleanly shutting them down.
I *like* the USB master port. I think the terapin is ridiculously priced (particularly when compared to the Archos Recorder 20 - USB 2.0, smaller form factor, 20Gb,lighter and plays *and* records mp3s).
The USB master port is a killer feature though - if it works as expected. In principle that should allow it to be directly connected to any slave device (eg: a digital camera) and data transferred. Seems like it would be extremely handy for backing up PDAs, and digital cameras.
Actually, it looks like device to device transfer will be built into some newer USB 2.0 devices. I would say slave to slave, but it looks like it is basically an integrated master/slave controller chipset that will be available:1 06S0058
http://www.eetimes.com/story/OEG20011
It will be called On-the-Go and it is specifically to work around this problem. Sounds great to me, I'd just love to have a Digital camera and be able to dump it to something like an Archos mp3/USB hdd.
My 5260 has a telnet interface as a previous poster described - you just set your interface (or a virtual interface if you use an OS capable of that) on your ethernet card to 10.0.0.2 and then telnet to 10.0.0.1.
The 5260 is a bridge (ethernet to err.. RADSL I think), so just because it is a bridge doens't mean it doesn't have that telnet interface.
Obviously if your OS is not capable of virtual interfaces so you can only have 1 IP per interface, then when you have your interface set to 10.0.0.2 you're not going to be able to get *past* the bridge, but that's okay, it is only temporary.
Also - the 5260 can be upgraded via firmware (it is out there somewhere) to be a router instead of a bridge. Neat, huh? Try www.dslreports.com to get a lot of info about these.
I'm going to hold onto my hard-earned cash until one comes out with GPS and MP3 capabilities also.
Preferably one that allows the mp3's to be stored on a removable hard drive.
Oh yeah, and a headset jack.
Oops.. I found the URL to it in their online catalog. Looks like they don't have them in the online catalog at the moment, so call around!
F na me=CTLG&category%5Fname=CTLG%5F002%5F003%5F001%5F0 00&product%5Fid=32%2D2010
http://www.radioshack.com/product.asp?catalog%5
Try good old radio shack. I was just in there today, and until the 28th of July they have a sale on a nice unit with wheels, a sloped top, and a straight up/down bottom. I don't know how many units, but pretty small. It looked to be plenty sturdy (built for amplifiers)
Price was $59 on sale. They also had a bigger one which was still less than $100.
The drawback is that they only have 1 small one per store, if that. So if you don't get the in store one you have to order it from the catalog, and I imagine shipping isn't cheap.
Link saturation seems to be the point -- on the principle that - "I've already paid for all this bandwidth, so why can't I use it to the fullest extent. If not getting something right now, then getting something that I might be needing".
It actually seems like a neat idea to me, and one that I've thought of (although if it became commonplace, I'll be ISP's would be failing left and right) as being useful.
Another useful thing would be to be able to "queue" up downloads explicitly (instead of implicitly by reference) (I think you can with some napster clients, although not in the ideal way) so that when the link is non-busy it will start sucking down data.
The hardest part to implement I think would be to ensure that your link isn't fully saturated when you want to use it. Then you would see a delayed response as your clever "look-ahead" backed down -- would you really see a net positive speed gain then?
The algorithm would have to be something like:
Is there any traffic currently, or has there been in the last X minutes? If so, then only use
half or 3/4 of what is currently available (ie: left over from what is currently being used, if any is).
else if during non-standard usage hours
use 9/10ths of the link
else if during standard usage hours
use 3/4 of the link
Obviously there would be a CPU price to pay, but if you've already got a much more powerful firewall/proxy than you need, why not?
Can you get that book (or others) at your public library? Sure, you might have to wait a week or two, but it usually can be done. If you're a college student, frequently college libraries are linked on a national level (while public libraries are usually linked locally) and you can order books from almost anywhere.
Information wants to be free.... or.. something.
I don't know if it would work, but I would certainly contribute.
The trouble is that the information is diverse and scattered -- how do you judge if a company is good or not? You could probably figure out a good rating system eg:
lowest - explicitly (stated) closed, and no drivers for alt OS
to highest - explicitly (stated) open, perhaps with patent, perhaps not, with open soucre developed w/ company time/money drivers.
Seems like it would be kinda worthless if it was all hearsay and the best thing would be to document each "point" eg: they *say* they are open source, here is a copy of the company mission statement. or: they produce freeBSD drivers, here is the link to the ftp site.
It would be great to be in a position to make not only compatability choices (x or y works well with Linux -- there are plenty of sites that will tell you that info) but choices based on the "moral/political" aspect of the company.
If it is designed as a site where users can add or remove points from a company by submitting documentation or links or something.. should work out pretty well.
I think "we" should spend a little less time posting our verbal spew on Slashdot and making LINUX more commercially ready. Or maybe not. Maybe "we" should spend our time how we want.
Maybe *you* are not the same as "we." You are lumping yourself in with this guy who did a port to dreamcast, and what have you yourself done besides this worthless post to Slashdot? Hopefully something. But if not, *I* don't care. Because *I* am spending my time posting a worthless followup to your waste of energy post. Because that is what *I* want to do.
And I should think that is exactly why this port to dreamcast was created. Because it was what he wanted to do.
So if *you* _think_ at all, and you _think_ that time should be put into making LINUX commercially viable (maybe, maybe not. That's not what Linus had in mind when he made it after all. Or maybe I don't know what Linus had in mind) then *you* will have to be the one to do it.
A fax connection is data anyway. Making a fax connection over an analog (voice) phone line was non-optimal in the first place. Why make it more complicated by doing data (fax) over voice (VoIP) over data (IP)? That's just plain ridiculous.
There are already fax over IP solutions available which work just fine. It would be stupid to do anything else.
Okay, but the announcement specifically talks about DVD playback. So if the hardware only accelerates MPEG then the software must take care of CSS. Presumably the closed-source portion of the software does this. So my question still remains -- will the software enforce zones or not?
It has been previously mentioned that many computer-based DVD solutions do not even enforce zones, it is just not publicitized, and of course a number of dedicated DVD players can be hacked so as not to enforce zones.
Does the software/hardware enforce zones, or is it zone-free? The press release makes no mention of this that I saw other than to say that the driver is released binary only for DVD security (probably required by their DVD licensing agreement) while the playback software will be open source (why couldn't they just make a plugin for xanim?).
And of course the other real question is was it a _Cuban_ cigar?
That may be the case, if so it is pretty neat and it is a very cheap device to have a custom drive in it. If that is the case though you should be able to easily do that simply using 2 drives (one reading while the other writes). That might actually end up being more expensive (particularly if you used larger drives, like I would), but that's not really the issue -- the issue is that is fun to mess around. :-)
I realize that Tivo is also partly selling a service which provides listings of all the shows and playtimes, etc which I would be more than happy to pay for independantly.
The real thing is being able to set up/mess around with your own hardware and software.
I'm a geek. This is great for a non-geek, and maybe many geeks would find it appealing as well.
But as a geek I'm very happy to use or configure a general purpose computer to do all of this stuff.
My most likely scenario would be one general-purpose computer _dedicated_ to all of the following, and one general-purpose computer for
general use (web surfing, word processing, etc).
The dedicated one would do:
dvd
tv/vcr (a-la Tivo if only the software were there!)
internet radio (preferably using dsp rather than cpu)
local mp3 storage
file server(?)
As a geek I see no need to buy a dedicated device when I can set up an even better dedicated device myself (and have fun doing it).
The place where I see myself buying a dedicated device where my geekly skills aren't quite up to making my own is a portable device for the car/motorcycle that does (or has modules for) all of the following:
gps w/uploadable mapping
packet radio location system (why not!)
mp3 (using a dsp, not cpu again)
radar detector -- using dsp algorithms like the valentine one, maybe with multiple antennas, etc so that you can set up and mess with the software settings (hopefully even source code) yourself.
weatherproof (for the motorcycle)
low power (yay transmeta!)
All of these things in a small box that plugs into the car or motorcycle would be ideal.
It's inefficient to do any of these for more than one OS...
...if it is difficult to do so for more than one OS, otherwise why not? If Java were really there...
Maybe it is just a hole in the market for a "VGM" -- virtual gaming machine that is coded platform independant from the OpenGL/sound drivers up to a nice easy gaming API. Automatic updates can then happen on the frontside of the API, and the VGM would only need to be updated very infrequently.
I don't know if this comparison has been made before but it seems to me that the Open Source movement is much like the Homebrew Computer Club.
The ideals in the case of the Homebrew Computer Club (as described in the interview) were: "empowering people without the companies owning the computers" only in this case it is software. We are in a better situation today -- almost anyone can be "hardware empowered," but if you want to be software empowered -- forget it.
Keep in mind that when you buy a piece of software like Windows or Office you are actually just buying a "license to use" the software. The software itself is *owned* my Microsoft (or Large Company of your Choice). You don't own it any more than people in the 70's owned the large computers that they rented/begged time on. Closed source is just one way that this non-ownership and disempowerment is enforced.
Want to get around the ATM fee? It's easy but the catch is that you usually have to go into a bank (or some institution) to do it.
Your checking/money market account should come with an associated debit card. This uses the standard VISA network. Simply make take a cash advance on your "VISA" card. There is no charge for cash advances by any insitution that I am aware of, although there is usually a cap on the amount, it is also usually much higher than you could get from an ATM (or the same, $500), plus you can get it in the bills of your choice! What could be better?
I guess that using the VISA network only costs the end bank some tiny fee and they just don't care about it, and they figure your bank is going to ream you with charges for taking a cash advance (they usually do) on your credit card. There's no need to inform them that it is not a credit card -- as far as they are concerned, it is. In one place I told them it was a debit card and they didn't care. So.. have fun.