The incentive to make the software is to solve a problem. Writing software shouldn't be a company's direct source of revenue. Creating solutions for their customers should be. All IP should be public domain by law, then the competition will be to sell services and create solutions to customer problems.
Companys have problems that need to be solved, they hire a programmer to create a solution. Its now in the public domain so there competition has it too, so to keep their competitive edge they need to focus on better service.
Yes, this kills off 1000's of jobs for programmers, most of whom shouldn't be programming anyway, they just went into the field because they heard there was lots of jobs and good money. Anyway programmers don't make any money off the software they write, uper-management and the investors make all the money.
This works in the music industry to, music that is created with money as an incentive sucks. The best music comes from love and pain, not from living it up with all the women you can handle. I'll pay $20 to go see a band play live, but I won't pay $20 for a CD just to hear the same songs over and over again.
Anyway this is the real world, so I'll just keep on downloading and pretending that all IP is public domain.
>Just to take a few of your assertations one at a time:
>> ask someone what was the last web ad they seen? > Can't remember. Must have been a few weeks ago thanks to a good adblocker. I'm talking about average joe here, most people don't use ad blockers... Hell I don't even bother with ad blockers, I just ignore most ads.
>> now ask someone what was the last tv commerical? >Can't remember. I use ad breaks to go to the toilet, make a cup of tea etc.
The few times I've watched TV in the past few months I've found the ads to be entertaining, they do get old and annoying really quickly, but the first time around they are entertaining.
>> Video ads usually get your attention with something entertaining... Some nice eye candy... > So I assume there are no boring ads on TV?
No but there are lots of entertaining ones out there...
>> Maybe advertizers will start demanding better content? > Advertiser: "Oh, look, new media format. Suddenly I can see the light!" Video ads cost alot more to produce,
> But finally, here's the one that really got me.
>> this might be the push we need to have average joe buy a new computer and broadband... > I live in central Tanzania. "Broadband" here costs $3,500/month, and runs at 312K. The average national > wage is about $400. "Average" may not mean exactly what you think it means.
Ofcourse broadband depends on many factors, the main one being demand, small demand and no supply == high cost, large demand and supply leads to competition and over subscription which means reasonable pricing.
Currently there is a lack of demand in the US for broadband, many average joe types a happy with dialup.
Also video ads don't stop people on dialup and on slow computers from viewing existing content.
Video ads are likely to be a more effective method of advertizing but advertisers are not going to put money into producing videos and bandwidth if there is no content to force people to view them. I think the name of the game is going to be producing websites that you view long enough to watch the entire video. So this means real content! Its not going to be a game of trying to get people to click the ads, but trying to get people to stay on one page long enough for the video to play to end. Now there will still be lots of annoying sites that try and force you to watch a video before accessing the content.
The fact that google is getting behind this is what I like. There are already companys producing video ads, I see these mircosoft ads here on slashdot all the time. Thankfully the sound is muted by default.
I personally don't mind the ads playing by default aslong as the sound is turned off. On my computers that are too slow to play video I simply don't have flash installed.
Broadband is widely avaiable here in Canada, unless you live in the middle of no where there is both adsl and cable.
There is a major lack of good quality legal broadband content online right now. Google video and yourtube are improving the situaltion but there is still very little professional quality content. Why? because there is no ad revenue to support it. Traditional banner ads or google ads barely cover the bandwidth if that. Hopefully video ads will pump some much needed money into this area.
Video based ads really get your attention, they are usually more entertaining that just a flashy logo and the companys latest deal that we see in flash ads.
Why is this a good thing? Everyone who visits a site will actually see the ad rather than ignore it, so webmasters can demand more money and stop producing websites that try and trick you into clicking an ads and stuff.
Video is a proven model or advertizing, ask someone what was the last web ad they seen? now ask someone what was the last tv commerical?
Maybe advertizers will start demanding better content?
Maybe we will see a whole new industry built on creating video ads for the web?
Video ads usually get your attention with something entertaining... Some nice eye candy...
This will really put a push on building better networks and pushing broadband, maybe advertizers will start putting pressure on broadband providers... Demand for bandwidth with money behind it, unlike bittorrent which is a demand for bandwidth without any money behind it.
Ofcourse there is the bad side, it dose suck if your on dialup or slow broadband, or have a slower computer, but all you have to do is uninstall flash or not install it in the first place, hell this might be the push we need to have average joe buy a new computer and broadband...
If there is enough demand there will be service. DSL services can be offered by installing a WIC(Walk in Cabinet) to house the DSLAM close to the subscribers. They use them all over the place here.
Basicly they have to have enough subscribers within an area to make the investment in fiber and the WIC worth while. The equipment to light up the fiber and the DSLAM's are a drop in the bucket compared to the cost of installing the fiber and WIC.
Your only options are to move or have your own fiber installed. The latter may not be so bad if you got a number of neighboors willing to share the bill. In that case I'd say the hell with the telco and also provide VoIP services.
It dosn't even have to be fiber! ADSL can do up to 8mbit which is more than enough for a few mpeg 4 streams.
Fiber rollout is very expensive, but its naturally going to happen over time. Your still going to require atleast one copper pair for 911 service, fiber cannot carry power and 911 services have to just work, even when the power is out.
HDTV is going to be a niche market for many years yet. I'm big into video and mpeg 4 on my regular TV is fine for me.
In the mean while someone needs to roll out a ADSL modem built into a set top box with an mpeg 4 decoder. Then the ISP's need to start providing multicast content for their subscribers. It could be done unicast too, on demand content is going to get pretty big. I've streamed video off my private colocated server to 1.5mbit adsl. It was xvid encoded and I used standard http for transport, the buffer set to 6 seconds and it never drops out(provideded the orginal video was well encoded)
Well if they did that their profits would fall through the floor. Most users don't use much bandwidth at all. Also power users would just switch to their competiors and tell there friends and family not to use their service. Also the average Joe user will just use the internet less to avoid extra changes.
Power users are something like 1% of users and consume 99% of the bandwith. With proper traffic shaping the other 99% of users won't even notice the 1% of power users. Traffic shaping of bittorrent is tricky but not impossible. Some people are using the VoIP ports to avoid shaping, but all an ISP has todo is montior the packet count on that port, if the speed exceeds say 128kbits its not voice and they can clamp down on it. They can also look at how much bandwidth a user consumes over a month, if it exceeds a set amount then their packets get the lowest priorty.
Another thing is if power users had to pay by the GB for the bandwidth would be paying well over $200 a month. At that price you might as well get your own dedicated fiber installed and sell your excess bandwidth to your neighboors.
More demand means prices will climb, more money leads to more bandwidth and the excess will then cause prices to drop, the natural cycle of it all...
Right now we are still paying for all the fiber thats out there. The big cost with fiber is putting it in the ground, lighting it up dosn't really cost that much in comparison. Also new technologies using DWDM allows providers to get alot more bandwidth out of a given fiber.
Right now there is no money to be made in increasing the amount of bandwidth. So no one is building faster backbones.
At the $50 a month people are willing to pay for broadband each user is only really getting something like 100kbit per second of bandwidth if not alot less. Ofcouse they can burst much much higher but the average bandwidth is actually very low per customer.
In short the internet will never really run out of bandwidth, if there is demand and people are willing to pay it will get faster.
Thats pretty standard in CATV networks these days. At the cable headend the RF spectrum from 30Mhz to 860Mhz is digitized(I belive) and sent out on fiber to each neighborhood served. From there a pole mounted unit converts the optical signal back to RF. Also on the pole signals below 30Mhz are converted to optical and sent back to the headend, this is how your cable modem gets data back to the headed.
Sending signals over long distances on RF over coax requires amplifiers to boost the signal along the way. Also as the signal is amplifed alittle bit of noise it added and the signal gets degraded. Also for two way services like DOCSIS(cable modems) there is a limited amount of shared RF bandwidth. An RF coax is like a hub, if you use one big hub to cover a city there isn't going to be much bandwidth. Running fiber to get neighborhood is like installing a switch to connect a number of hubs and increases the amount of bandwidth. The smaller number of customers per hub the more bandwith avaible. So the smaller area the RF part of the network covers the better.
With a simple bit of traffic shaping a single 1.5mbit DSL line can handle the webbroswing of alot of users. Even if a few are using Bittorrent and other bandwidth hungry applications. The big downloads will be a little slow but one shouldn't notice latency on small transfers if its setup properly.
The real problem with Bittorrent on ADSL is on the upload side. The send queue on the modem fills up and packets will take a few 100ms to get through if they don't get dropped. This makes for a painfully slow experience.
Greater than 1.5mbit service only really required when you want to offer services like streaming media. Using a MPEG4 codec like xvid you can stream fairly good quality TV at 1.5mbit but that dosn't leave much for overhead and other applications. Also thats only one channel, these days your typical home may have 4 people watching 4 different things on 4 differnt channels, so then you need 6mbit of bandwidth.
No major provider is going to get behind peer to peer. The idea behind p2p is to avoid the bandwidth cost. Well thats lost income for the provider. Peer to peer is a cool idea but in the long run its going to be squashed.
I can see the day comming when its impossible to get a publicly routed IPv4 address to your home. Some ISP's are already using private addressing for their subscribers. The switch to IPv6 just isn't happening and there really isn't a need. Between virtual hosting and NAT the IP address shortage has been solved. No desktop computer really needs an Internet routable IP and this also adds a layer of security.
As much as I love getting my weekly fix of TV for free off Bittorrent, I just don't see ISP's allowing this to continue for much longer. Once they work out an effective way to stream content I can see them filtering it out all together. They are not going to let people to get what they are selling for free. Both of the local broadband ISP's here already are cracking down on it by heavly throttling all traffic to users who exceed a cerntain threshold.
Very few public sites can supply a single user with 6mbit. Most servers are still on 100mbit ethernet and are serving alot more than 20 clients at a time. Currently the only way to get more than 1.5mbit from the public internet is via bittorrent. Unless you have your own server in a datacenter thats not seeing much load. When I had 5mbit DSL I could download at 500kB/s from my colocated server.
Linux, FreeBSD, MacOSX and Windows XP all just tools and some are better at some task than other. Windows XP is more than useable and will run on just about any x86 beyond a PPro, and in my experince its faster than linux on the desktop. On this little laptop for example, a PII366/w 128megs of ram, windows XP is slightly faster running apps like firefox and will play mpeg2 video in vlc while under linux it chokes. Windows on x86 hardware is better for multimedia and gaming.
Windows XP is usually slow because people install all kinds of spyware and other junk. In my experince antivirus software is worse than the malware its trying to keep out. It slows computers to a crawl. I've used windows on and off since win3.1 and I have never used antivirus software. I got one bootsector virus back in the day and the msblaster worm a few years ago. It just comes down to having a hardware firewall(well just NAT) and not using outlook, IE, word or hotmail. Basicly MS OS + MS APP = trouble !
Linux is great on server where you need the latest and greatest, *BSD where you need alittle more stablity and security. I'd never trust a closed OS wide open on the internet.
Personally I think Linux on the home desktop is a lost cause, there is no real advantage anymore. Windows XP is secure enough and has far better hardware suppport. Alot of the FOSS that was orginally written for linux usually runs on winxp, if not better on winxp. Firefox, thunderbird, vlc, ethereal, the gimp, etc... I'm yet to install linux on a machine and have everything just work. I'm also getting really sick of reading somewhere that something is suppored and then when I go to use it, its not finshed, missing features, unstable or just dosn't work at all.
You start putting linux on the desktop's of the general public and its going to run into all of the same problems you have under windows. Atleast under linux I'd hope the damage would be limited to the user's account but how many average joe's are not going to use the root account or give up their root password to a malware program promising free porn.
Now linux dose have a place on the desktop in big business, where you have someone to administer the systems. Diskless systems custom built with linux friendly hardware. All that is needed here is a polished up version of Open Office, firefox and thunderbird. I think we will see a web based solution the way things are going.
Well thats kinda the point of it being ajax... Ever use gmail? notice the autosave? thats ajax at work. You can use ajax so that every keystroke gets sent to the server and stored in an SQL database, but thats alittle wasteful of bandwidth.
I've had more problems losing data in word than online. My servers have various levels of redundancy, they are running in a datacentre with redundant power and with high physical security. Data loss would require two harddrives to fail at once or if it information hasn't been written to disk yet, two servers to lose poswer at the same time.
As much as I hate lying, for a job like that your not going to have the experince with the IVR software, people who have that experience are already working in that field. So lie... Do lots of research before heading into the interview and get up to speed on the technologies they are looking for.
If they are being unrealistic then you will have to be... Or someone else will be and take the job...
Any programmer should be able to jump languages without blinking, each one has its quirks but they are all pretty much the same.
I've also heard of people using friends as references to back up lies about experience.
I personally haven't done this, and I hope never to have to, but I have had friends with no real experince and no education get jobs requiring a CS degree and 5 years experince. He lied his ass off, pawed off work on others and made his way into management... He is one of these people we techies hate...
This seems to be alittle late to be jumping into this market. Most of the big players are starting to switch over to multilayer switching. Software routers are only needed where you need to do something like NAT or firewalling.
If your big enough to need a routing protocol like BGP, your going to need some serious hardware. Software based routers running on off the self hardware are fine for 100mbit ethernet routing, but beyond 100mbit you need some specialized hardware.
I really don't see any advanage this system has over a linux router with the usual tools(zebra/quagga, ip, ifconfig, iptables, ebtables, etc...)
I think the selling point is going to be no more data loss and no more malware. Most average consume types I talk to find that all of this security stuff is confusing and are tired of loosing files. Thin clients put all the hardwork in security and reliablity in the datacenter.
The whole thing has to be sold as one simple package for a reasonable price. The package must include internet, voice, media, office applications, personal webspace(blog, photo albums, shared documents, etc... all ofcouse intergrated into the desktop). The only company even close to offering this today is google and they only have the datacenter end of it down, google needs to start buying up telcos.
I don't think people will mind buying there own hardware, they are going to be buying on looks mainly, getting something that matches their desk. The hardware has to become standardized, like telephone sets. None of this Y won't work with X.
Also thin client is not the same as VNC or remote desktop, the client still dose alot of the desktop rendering and thus uses less bandwidth. I think we are going to see thin clients with a built in webbrowser that can do alot of the redering.
I don't think we will ever see average joe owning his own server. No matter how easy or cheap it gets servers are not a concept that the average person really gets and this isn't about to change in the next 15 years.
I think your more likely to see diskless systems and fiber to the home. Hell ADSL is fast enough to stream better than cable quality xvid encoded video now. The powers at be don't want people to own massive collections of media and what else do you need all the HD space for on a Personal Computer? A gig of flash will store the OS and all the applications needed, your data will be stored by your ISP or telco(I'd rather the later).
I think we are going to see wireless LANs become as common as cordless phones, its almost there already. Someone just has to come up with a method of secure auto configureation(WPA or better by default) and more RF bandwidth has to be made avaiable for wireless LAN. Average joe gets home with his new media center machine he is just going to want to plug in power and video and go. None of this messing around with running UTP or someday fiber. The service provider will supply and configure the CPE to terminate the fiber or DSL and could also provide wireless access.
I think alot of people will still have desktops, they won't be the greatest or the latest thing that people still go for today. The idea that you have to have the latest and the greatest is so 90's. Any computer sold in the past 5 years has more than enough CPU power and resources to handle any day to day task. Its only gaming and media production that require top end machines at home. I think we will see totally intergrated machines around the size of an lcd montior, if not smaller. Basicly a laptop without drives and input devices(well maybe a touch screen). Add wireless keyboard and mouse and away you go.
Consumers demand simple and easy, and indirectly secure and reliable. Alot of the more difficult things to do are going to be moved to the datacenter. Thats where us computer geeks will belong. The home computer will just be a terminal for all the software and media out there.
People might say what happends when you internet goes down, well when was the last time your phone line went down? As the technology matures, the reliabilty should go up. Also there will be still standalone laptops.
I can't see this being much worse than all the p2p and malware traffic out there right now. The only thing is that p2p dosn't usually use port 80 so its been easy for ISP to apply traffic shapeing to throttle p2p and keep the web fast.
We also have free local calling here in Canada and broadband is everywhere. Its only like $10 more expensive than dialup was and its been around for a while now. Its usually simpler and easier, have someone come in and set it up, and you never have to mess around to get online. No one seems to even consider dialup anymore.
Here ADSL is being offered by the ILEC rather than 3rd parties. There are competive ADSL in the major cities. Where its being offered by a telco, you get telco grade service, always fast, always up, no bandwidth limits. Also the telco are constantly upgrading the lines to carry adsl, keeping the system bridge tap free and installing remote terminals everywhere. The prices are pretty good too, an extra $35/month on the phone bill for 1.5/512, add another $5 to get 5/512. My only beef is that you have to have phone service on the line. My main phone is a cell. Right now I'm just using a neighboors connection via wireless.
There is also broadband cable in most areas offered by Rogers, which some people get suckered into. Its slow, unreliable, filtered(P2P dosn't work at all). The only good thing about it is that it keeps the pricing competitive and it fills in a few gaps where adsl is not avaiable yet.
ADSL have been avaible everywhere for the past 7 or 8 years. When I first got it, the basic package was like 512k/128k, I went for the best package 1024/768. The ILECs jumped right into offering adsl and never looked back here in Canada. Thats what its sucessful here. It might also have something todo with the condition of the cable plants when then started out.
Rogers here in St. John's, NL have somehow blocked Bittorrent all together, I think they are using some sort of high level filtering to block the connection to the tracker.
Aliant so far has been pretty fast with bittorrent, they now offer 5mbit as an upgrade from 1.5mbit for an extra $5 a month. Too bad they are keeping the upload capped at 512kbit, when I first signed up years ago it was 1mbit down, 768kbit up. My parents still had this service untill they had to "upgrade" when they switched to a bundled package a few months ago.
The difference in speed is more likely todo with other factors. Aliant's network is well over built to handle the load of their customers using Bittorrent. Bandwidth only really cost them money when it crosses over to the US. They are owned by bell, who also owns Group telecom... Here in St. John's that means their are only two choices Bell or Rogers.
When I was living in Halifax last fall I was maxing out my 5mbit connection on some well seeded torrents(private tracker). I also had no problem maxing out the 5mbit downloading over http from a colocated server I admin on rogers telecom's backbone in Alberta.
Last winter I had Rogers 5mbit here in St. John's, it was fast for the first few months, then it got slower and slower, then bittorrent just stopped working all together. In the end I was pretty fed up with Rogers, When I subscribed they told be their was no commitment, then when I went to cancel they said I needed to give them 30 day notice and their was going to be a cancelaltion fee. Also the monthly price went up 3 times, the first time was after the 3 month deal, it went from the really cheap $15 a month, to a reasonable $30 a month. A month or two later they started to charge me for basic cable(without removing the filter) so another $10 a month, then prices went up again and it was almost $50 a month.
My dealings with aliant were alot better, I signed up, got service as promissed after they had troble with shipping the modem to me and gave me free service for a month, the next month I paid and then canceled when I broke up eith my ex. No problem at all, no cancelation fees or anything. They did over charge me for the last month but when I called they wavied the fee for the last month because it was higher than it should of been.
Aliants network is always fast, reliable, and pretty cheap if you get their bundles. I had cell, landline, 5mbit adsl, unlimited atlantic canada longdistance and calling features for $85 a month. From rogers you would be luck to get cell and 5mbit internet for that.
The problem is that not everyone I know uses MSN, most people do but alot don't. One thing that pisses me off about MSN is that I have to add everyone manually or they have to add me. I'd rather it be more like IRC so I could contact people who I haven't seen in years. Alot of my friends still use the orginal IM, IRC, which is good for contacting people I haven't seen in a while. A few other people I know only use xfire, others use yahoo.
A HDD will last about 10 years with constant use, but just sitting on the shelf in a dry enviorment it should last pretty much forever. Optical media on the otherhand slowly reacts with the air and light.
Anyway the big problem with optical is that you can only store 4.7 gigs on a DVDR, which is nothing to this guy. HDD's and Tape are the only possible solutions for this guys problem. I'd go with two HD's on firewire or USB2.0 and storing atleast one of them off site at the end of the day. Tape can be ok too but what is the seek time like on todays tech? If he is looking for one clip is he going to have to ff through the whole tape?
The answer to me seems like some form of software raid setup for write once only to external HD's.
You kinda got that backwards, Desktops are becoming the niche. The average user wants something simple, one power cable and thats it. Thats exactly what laptops offer. When they do break its easy for them to just close the screen and carry to the nearest service depot.
I, on the other hand, just buy used laptops off ebay. No need for a 1Ghz+ machine for just running firefox and ssh. When something breaks I got a junker that I can grab part from. My current machine has been through hell and back and still runs great, I think I'm on my 3rd battery, I keep lucking in and getting pretty desent batteries with the used machines. Hell I can even run some old games on this machine. My current laptop is a dell latitude PII 400/w 256megs of ram, only got a 6.5gig HD, but thats where the lan comes in.
I still have a powerful desktop machine that I use for gaming and lately mostly just a media player. The laptop is more pratical for day to day use. I'm use to the keyboard, more so than my desktop's keyboard.
I can see desktops in their current modular/user upgradeable form becoming more expensive than laptops in a few years. Custom built desktops will become a niche for power users and gamers. While adverage users go for iMac like all in one systems and laptops, and business goes towards thin clients. Don't get me wrong, this isn't going to happen over night.
Power users use way more bandwidth than your average no-clue person. They would ofcourse charge what they charge now for level 0 and the higher levels would start atleast double. Thats captialism there for you. No ISP wants power users unless they can find a way to profit from them.
In general your going to have to find books on each specfic technologies. Most of the general ISP resources are from the mid 90's when dial up was all the rage.
An ISP sysadmin is a very specialized area and over the past few years the field has gotten smaller and smaller with everyone buying out everyone else. So your pretty much on your own.
I've just spent the past year doing Telecom Eng. Technology at the Collage of the North Atlantic. The courses included L2 to L4 switching, tcp/ip, network cabling, LAN's and DSL(Why they put these together), digital telephony, voip and prevoius to that I did a mess of eletronics and math courses. That broad range of courses still left out alot of topics which I covered on my own. I really made of the best of all the lab time I could get, built my own model internet out of old 2600 serise routers, a few linux boxes/w OC-3 cards, multilayer switches, DSLAM's and wireless bridges. Got a chance to mess around with a good mix of different layer 1 and layer 2 technologies along with a few different routing protocols. One course I totally underestimated was level 2 to 4 switching, I had no idea that ethernet switching was so complex.
Anyway my point is, you have to understand the underlying technologies and from there build a working knowledge of how to use your equipment at hand.
Oh yea and don't underestimate the power of linux in networking. A linux box can handle just about any task/service above layer 2, just gota get good hardware and with the money you save not buying cisco buy two or more, so you can always have a hot backup.
Slashdot Mirror
The incentive to make the software is to solve a problem. Writing software shouldn't be a company's direct source of revenue. Creating solutions for their customers should be. All IP should be public domain by law, then the competition will be to sell services and create solutions to customer problems.
Companys have problems that need to be solved, they hire a programmer to create a solution. Its now in the public domain so there competition has it too, so to keep their competitive edge they need to focus on better service.
Yes, this kills off 1000's of jobs for programmers, most of whom shouldn't be programming anyway, they just went into the field because they heard there was lots of jobs and good money. Anyway programmers don't make any money off the software they write, uper-management and the investors make all the money.
This works in the music industry to, music that is created with money as an incentive sucks. The best music comes from love and pain, not from living it up with all the women you can handle. I'll pay $20 to go see a band play live, but I won't pay $20 for a CD just to hear the same songs over and over again.
Anyway this is the real world, so I'll just keep on downloading and pretending that all IP is public domain.
>Just to take a few of your assertations one at a time:
>> ask someone what was the last web ad they seen?
> Can't remember. Must have been a few weeks ago thanks to a good adblocker.
I'm talking about average joe here, most people don't use ad blockers... Hell I don't even bother with ad blockers, I just ignore most ads.
>> now ask someone what was the last tv commerical?
>Can't remember. I use ad breaks to go to the toilet, make a cup of tea etc.
The few times I've watched TV in the past few months I've found the ads to be entertaining, they do get old and annoying really quickly, but the first time around they are entertaining.
>> Video ads usually get your attention with something entertaining... Some nice eye candy...
> So I assume there are no boring ads on TV?
No but there are lots of entertaining ones out there...
>> Maybe advertizers will start demanding better content?
> Advertiser: "Oh, look, new media format. Suddenly I can see the light!"
Video ads cost alot more to produce,
> But finally, here's the one that really got me.
>> this might be the push we need to have average joe buy a new computer and broadband...
> I live in central Tanzania. "Broadband" here costs $3,500/month, and runs at 312K. The average national > wage is about $400. "Average" may not mean exactly what you think it means.
Ofcourse broadband depends on many factors, the main one being demand, small demand and no supply == high cost, large demand and supply leads to competition and over subscription which means reasonable pricing.
Currently there is a lack of demand in the US for broadband, many average joe types a happy with dialup.
Also video ads don't stop people on dialup and on slow computers from viewing existing content.
Video ads are likely to be a more effective method of advertizing but advertisers are not going to put money into producing videos and bandwidth if there is no content to force people to view them. I think the name of the game is going to be producing websites that you view long enough to watch the entire video. So this means real content! Its not going to be a game of trying to get people to click the ads, but trying to get people to stay on one page long enough for the video to play to end. Now there will still be lots of annoying sites that try and force you to watch a video before accessing the content.
The fact that google is getting behind this is what I like. There are already companys producing video ads, I see these mircosoft ads here on slashdot all the time. Thankfully the sound is muted by default.
I personally don't mind the ads playing by default aslong as the sound is turned off. On my computers that are too slow to play video I simply don't have flash installed.
Broadband is widely avaiable here in Canada, unless you live in the middle of no where there is both adsl and cable.
There is a major lack of good quality legal broadband content online right now. Google video and yourtube are improving the situaltion but there is still very little professional quality content. Why? because there is no ad revenue to support it. Traditional banner ads or google ads barely cover the bandwidth if that. Hopefully video ads will pump some much needed money into this area.
Video based ads really get your attention, they are usually more entertaining that just a flashy logo and the companys latest deal that we see in flash ads.
Why is this a good thing? Everyone who visits a site will actually see the ad rather than ignore it, so webmasters can demand more money and stop producing websites that try and trick you into clicking an ads and stuff.
Video is a proven model or advertizing, ask someone what was the last web ad they seen? now ask someone what was the last tv commerical?
Maybe advertizers will start demanding better content?
Maybe we will see a whole new industry built on creating video ads for the web?
Video ads usually get your attention with something entertaining... Some nice eye candy...
This will really put a push on building better networks and pushing broadband, maybe advertizers will start putting pressure on broadband providers... Demand for bandwidth with money behind it, unlike bittorrent which is a demand for bandwidth without any money behind it.
Ofcourse there is the bad side, it dose suck if your on dialup or slow broadband, or have a slower computer, but all you have to do is uninstall flash or not install it in the first place, hell this might be the push we need to have average joe buy a new computer and broadband...
If there is enough demand there will be service. DSL services can be offered by installing a WIC(Walk in Cabinet) to house the DSLAM close to the subscribers. They use them all over the place here.
Basicly they have to have enough subscribers within an area to make the investment in fiber and the WIC worth while. The equipment to light up the fiber and the DSLAM's are a drop in the bucket compared to the cost of installing the fiber and WIC.
Your only options are to move or have your own fiber installed. The latter may not be so bad if you got a number of neighboors willing to share the bill. In that case I'd say the hell with the telco and also provide VoIP services.
It dosn't even have to be fiber! ADSL can do up to 8mbit which is more than enough for a few mpeg 4 streams.
Fiber rollout is very expensive, but its naturally going to happen over time. Your still going to require atleast one copper pair for 911 service, fiber cannot carry power and 911 services have to just work, even when the power is out.
HDTV is going to be a niche market for many years yet. I'm big into video and mpeg 4 on my regular TV is fine for me.
In the mean while someone needs to roll out a ADSL modem built into a set top box with an mpeg 4 decoder. Then the ISP's need to start providing multicast content for their subscribers. It could be done unicast too, on demand content is going to get pretty big. I've streamed video off my private colocated server to 1.5mbit adsl. It was xvid encoded and I used standard http for transport, the buffer set to 6 seconds and it never drops out(provideded the orginal video was well encoded)
Well if they did that their profits would fall through the floor. Most users don't use much bandwidth at all. Also power users would just switch to their competiors and tell there friends and family not to use their service. Also the average Joe user will just use the internet less to avoid extra changes.
Power users are something like 1% of users and consume 99% of the bandwith. With proper traffic shaping the other 99% of users won't even notice the 1% of power users. Traffic shaping of bittorrent is tricky but not impossible. Some people are using the VoIP ports to avoid shaping, but all an ISP has todo is montior the packet count on that port, if the speed exceeds say 128kbits its not voice and they can clamp down on it. They can also look at how much bandwidth a user consumes over a month, if it exceeds a set amount then their packets get the lowest priorty.
Another thing is if power users had to pay by the GB for the bandwidth would be paying well over $200 a month. At that price you might as well get your own dedicated fiber installed and sell your excess bandwidth to your neighboors.
More demand means prices will climb, more money leads to more bandwidth and the excess will then cause prices to drop, the natural cycle of it all...
Right now we are still paying for all the fiber thats out there. The big cost with fiber is putting it in the ground, lighting it up dosn't really cost that much in comparison. Also new technologies using DWDM allows providers to get alot more bandwidth out of a given fiber.
Right now there is no money to be made in increasing the amount of bandwidth. So no one is building faster backbones.
At the $50 a month people are willing to pay for broadband each user is only really getting something like 100kbit per second of bandwidth if not alot less. Ofcouse they can burst much much higher but the average bandwidth is actually very low per customer.
In short the internet will never really run out of bandwidth, if there is demand and people are willing to pay it will get faster.
Thats pretty standard in CATV networks these days. At the cable headend the RF spectrum from 30Mhz to 860Mhz is digitized(I belive) and sent out on fiber to each neighborhood served. From there a pole mounted unit converts the optical signal back to RF. Also on the pole signals below 30Mhz are converted to optical and sent back to the headend, this is how your cable modem gets data back to the headed.
Sending signals over long distances on RF over coax requires amplifiers to boost the signal along the way. Also as the signal is amplifed alittle bit of noise it added and the signal gets degraded. Also for two way services like DOCSIS(cable modems) there is a limited amount of shared RF bandwidth. An RF coax is like a hub, if you use one big hub to cover a city there isn't going to be much bandwidth. Running fiber to get neighborhood is like installing a switch to connect a number of hubs and increases the amount of bandwidth. The smaller number of customers per hub the more bandwith avaible. So the smaller area the RF part of the network covers the better.
With a simple bit of traffic shaping a single 1.5mbit DSL line can handle the webbroswing of alot of users. Even if a few are using Bittorrent and other bandwidth hungry applications. The big downloads will be a little slow but one shouldn't notice latency on small transfers if its setup properly.
The real problem with Bittorrent on ADSL is on the upload side. The send queue on the modem fills up and packets will take a few 100ms to get through if they don't get dropped. This makes for a painfully slow experience.
Greater than 1.5mbit service only really required when you want to offer services like streaming media. Using a MPEG4 codec like xvid you can stream fairly good quality TV at 1.5mbit but that dosn't leave much for overhead and other applications. Also thats only one channel, these days your typical home may have 4 people watching 4 different things on 4 differnt channels, so then you need 6mbit of bandwidth.
No major provider is going to get behind peer to peer. The idea behind p2p is to avoid the bandwidth cost. Well thats lost income for the provider. Peer to peer is a cool idea but in the long run its going to be squashed.
I can see the day comming when its impossible to get a publicly routed IPv4 address to your home. Some ISP's are already using private addressing for their subscribers. The switch to IPv6 just isn't happening and there really isn't a need. Between virtual hosting and NAT the IP address shortage has been solved. No desktop computer really needs an Internet routable IP and this also adds a layer of security.
As much as I love getting my weekly fix of TV for free off Bittorrent, I just don't see ISP's allowing this to continue for much longer. Once they work out an effective way to stream content I can see them filtering it out all together. They are not going to let people to get what they are selling for free. Both of the local broadband ISP's here already are cracking down on it by heavly throttling all traffic to users who exceed a cerntain threshold.
Very few public sites can supply a single user with 6mbit. Most servers are still on 100mbit ethernet and are serving alot more than 20 clients at a time. Currently the only way to get more than 1.5mbit from the public internet is via bittorrent. Unless you have your own server in a datacenter thats not seeing much load. When I had 5mbit DSL I could download at 500kB/s from my colocated server.
Linux, FreeBSD, MacOSX and Windows XP all just tools and some are better at some task than other. Windows XP is more than useable and will run on just about any x86 beyond a PPro, and in my experince its faster than linux on the desktop. On this little laptop for example, a PII366 /w 128megs of ram, windows XP is slightly faster running apps like firefox and will play mpeg2 video in vlc while under linux it chokes. Windows on x86 hardware is better for multimedia and gaming.
Windows XP is usually slow because people install all kinds of spyware and other junk. In my experince antivirus software is worse than the malware its trying to keep out. It slows computers to a crawl. I've used windows on and off since win3.1 and I have never used antivirus software. I got one bootsector virus back in the day and the msblaster worm a few years ago. It just comes down to having a hardware firewall(well just NAT) and not using outlook, IE, word or hotmail. Basicly MS OS + MS APP = trouble !
Linux is great on server where you need the latest and greatest, *BSD where you need alittle more stablity and security. I'd never trust a closed OS wide open on the internet.
Personally I think Linux on the home desktop is a lost cause, there is no real advantage anymore. Windows XP is secure enough and has far better hardware suppport. Alot of the FOSS that was orginally written for linux usually runs on winxp, if not better on winxp. Firefox, thunderbird, vlc, ethereal, the gimp, etc... I'm yet to install linux on a machine and have everything just work. I'm also getting really sick of reading somewhere that something is suppored and then when I go to use it, its not finshed, missing features, unstable or just dosn't work at all.
You start putting linux on the desktop's of the general public and its going to run into all of the same problems you have under windows. Atleast under linux I'd hope the damage would be limited to the user's account but how many average joe's are not going to use the root account or give up their root password to a malware program promising free porn.
Now linux dose have a place on the desktop in big business, where you have someone to administer the systems. Diskless systems custom built with linux friendly hardware. All that is needed here is a polished up version of Open Office, firefox and thunderbird. I think we will see a web based solution the way things are going.
Well thats kinda the point of it being ajax... Ever use gmail? notice the autosave? thats ajax at work. You can use ajax so that every keystroke gets sent to the server and stored in an SQL database, but thats alittle wasteful of bandwidth.
I've had more problems losing data in word than online. My servers have various levels of redundancy, they are running in a datacentre with redundant power and with high physical security. Data loss would require two harddrives to fail at once or if it information hasn't been written to disk yet, two servers to lose poswer at the same time.
Those requirements are pretty realistic...
As much as I hate lying, for a job like that your not going to have the experince with the IVR software, people who have that experience are already working in that field. So lie... Do lots of research before heading into the interview and get up to speed on the technologies they are looking for.
If they are being unrealistic then you will have to be... Or someone else will be and take the job...
Any programmer should be able to jump languages without blinking, each one has its quirks but they are all pretty much the same.
I've also heard of people using friends as references to back up lies about experience.
I personally haven't done this, and I hope never to have to, but I have had friends with no real experince and no education get jobs requiring a CS degree and 5 years experince. He lied his ass off, pawed off work on others and made his way into management... He is one of these people we techies hate...
This seems to be alittle late to be jumping into this market. Most of the big players are starting to switch over to multilayer switching. Software routers are only needed where you need to do something like NAT or firewalling.
If your big enough to need a routing protocol like BGP, your going to need some serious hardware. Software based routers running on off the self hardware are fine for 100mbit ethernet routing, but beyond 100mbit you need some specialized hardware.
I really don't see any advanage this system has over a linux router with the usual tools(zebra/quagga, ip, ifconfig, iptables, ebtables, etc...)
Why MPEG 2 and not MPEG 4?
I think the selling point is going to be no more data loss and no more malware. Most average consume types I talk to find that all of this security stuff is confusing and are tired of loosing files. Thin clients put all the hardwork in security and reliablity in the datacenter.
The whole thing has to be sold as one simple package for a reasonable price. The package must include internet, voice, media, office applications, personal webspace(blog, photo albums, shared documents, etc... all ofcouse intergrated into the desktop). The only company even close to offering this today is google and they only have the datacenter end of it down, google needs to start buying up telcos.
I don't think people will mind buying there own hardware, they are going to be buying on looks mainly, getting something that matches their desk. The hardware has to become standardized, like telephone sets. None of this Y won't work with X.
Also thin client is not the same as VNC or remote desktop, the client still dose alot of the desktop rendering and thus uses less bandwidth. I think we are going to see thin clients with a built in webbrowser that can do alot of the redering.
I don't think we will ever see average joe owning his own server. No matter how easy or cheap it gets servers are not a concept that the average person really gets and this isn't about to change in the next 15 years.
I think your more likely to see diskless systems and fiber to the home. Hell ADSL is fast enough to stream better than cable quality xvid encoded video now. The powers at be don't want people to own massive collections of media and what else do you need all the HD space for on a Personal Computer? A gig of flash will store the OS and all the applications needed, your data will be stored by your ISP or telco(I'd rather the later).
I think we are going to see wireless LANs become as common as cordless phones, its almost there already. Someone just has to come up with a method of secure auto configureation(WPA or better by default) and more RF bandwidth has to be made avaiable for wireless LAN. Average joe gets home with his new media center machine he is just going to want to plug in power and video and go. None of this messing around with running UTP or someday fiber. The service provider will supply and configure the CPE to terminate the fiber or DSL and could also provide wireless access.
I think alot of people will still have desktops, they won't be the greatest or the latest thing that people still go for today. The idea that you have to have the latest and the greatest is so 90's. Any computer sold in the past 5 years has more than enough CPU power and resources to handle any day to day task. Its only gaming and media production that require top end machines at home. I think we will see totally intergrated machines around the size of an lcd montior, if not smaller. Basicly a laptop without drives and input devices(well maybe a touch screen). Add wireless keyboard and mouse and away you go.
Consumers demand simple and easy, and indirectly secure and reliable. Alot of the more difficult things to do are going to be moved to the datacenter. Thats where us computer geeks will belong. The home computer will just be a terminal for all the software and media out there.
People might say what happends when you internet goes down, well when was the last time your phone line went down? As the technology matures, the reliabilty should go up. Also there will be still standalone laptops.
I can't see this being much worse than all the p2p and malware traffic out there right now. The only thing is that p2p dosn't usually use port 80 so its been easy for ISP to apply traffic shapeing to throttle p2p and keep the web fast.
We also have free local calling here in Canada and broadband is everywhere. Its only like $10 more expensive than dialup was and its been around for a while now. Its usually simpler and easier, have someone come in and set it up, and you never have to mess around to get online. No one seems to even consider dialup anymore.
Here ADSL is being offered by the ILEC rather than 3rd parties. There are competive ADSL in the major cities. Where its being offered by a telco, you get telco grade service, always fast, always up, no bandwidth limits. Also the telco are constantly upgrading the lines to carry adsl, keeping the system bridge tap free and installing remote terminals everywhere. The prices are pretty good too, an extra $35/month on the phone bill for 1.5/512, add another $5 to get 5/512. My only beef is that you have to have phone service on the line. My main phone is a cell. Right now I'm just using a neighboors connection via wireless.
There is also broadband cable in most areas offered by Rogers, which some people get suckered into. Its slow, unreliable, filtered(P2P dosn't work at all). The only good thing about it is that it keeps the pricing competitive and it fills in a few gaps where adsl is not avaiable yet.
ADSL have been avaible everywhere for the past 7 or 8 years. When I first got it, the basic package was like 512k/128k, I went for the best package 1024/768. The ILECs jumped right into offering adsl and never looked back here in Canada. Thats what its sucessful here. It might also have something todo with the condition of the cable plants when then started out.
Rogers here in St. John's, NL have somehow blocked Bittorrent all together, I think they are using some sort of high level filtering to block the connection to the tracker.
Aliant so far has been pretty fast with bittorrent, they now offer 5mbit as an upgrade from 1.5mbit for an extra $5 a month. Too bad they are keeping the upload capped at 512kbit, when I first signed up years ago it was 1mbit down, 768kbit up. My parents still had this service untill they had to "upgrade" when they switched to a bundled package a few months ago.
The difference in speed is more likely todo with other factors. Aliant's network is well over built to handle the load of their customers using Bittorrent. Bandwidth only really cost them money when it crosses over to the US. They are owned by bell, who also owns Group telecom... Here in St. John's that means their are only two choices Bell or Rogers.
When I was living in Halifax last fall I was maxing out my 5mbit connection on some well seeded torrents(private tracker). I also had no problem maxing out the 5mbit downloading over http from a colocated server I admin on rogers telecom's backbone in Alberta.
Last winter I had Rogers 5mbit here in St. John's, it was fast for the first few months, then it got slower and slower, then bittorrent just stopped working all together. In the end I was pretty fed up with Rogers, When I subscribed they told be their was no commitment, then when I went to cancel they said I needed to give them 30 day notice and their was going to be a cancelaltion fee. Also the monthly price went up 3 times, the first time was after the 3 month deal, it went from the really cheap $15 a month, to a reasonable $30 a month. A month or two later they started to charge me for basic cable(without removing the filter) so another $10 a month, then prices went up again and it was almost $50 a month.
My dealings with aliant were alot better, I signed up, got service as promissed after they had troble with shipping the modem to me and gave me free service for a month, the next month I paid and then canceled when I broke up eith my ex. No problem at all, no cancelation fees or anything. They did over charge me for the last month but when I called they wavied the fee for the last month because it was higher than it should of been.
Aliants network is always fast, reliable, and pretty cheap if you get their bundles. I had cell, landline, 5mbit adsl, unlimited atlantic canada longdistance and calling features for $85 a month. From rogers you would be luck to get cell and 5mbit internet for that.
I think its more the other way around....
Too much and it make life bland and tasteless.
Too little will make you die.
Why don't they build the datacenter up north, use the heat gernerated to heat water and pump it all over town?
The problem is that not everyone I know uses MSN, most people do but alot don't. One thing that pisses me off about MSN is that I have to add everyone manually or they have to add me. I'd rather it be more like IRC so I could contact people who I haven't seen in years. Alot of my friends still use the orginal IM, IRC, which is good for contacting people I haven't seen in a while. A few other people I know only use xfire, others use yahoo.
A HDD will last about 10 years with constant use, but just sitting on the shelf in a dry enviorment it should last pretty much forever. Optical media on the otherhand slowly reacts with the air and light.
Anyway the big problem with optical is that you can only store 4.7 gigs on a DVDR, which is nothing to this guy. HDD's and Tape are the only possible solutions for this guys problem. I'd go with two HD's on firewire or USB2.0 and storing atleast one of them off site at the end of the day. Tape can be ok too but what is the seek time like on todays tech? If he is looking for one clip is he going to have to ff through the whole tape?
The answer to me seems like some form of software raid setup for write once only to external HD's.
You kinda got that backwards, Desktops are becoming the niche. The average user wants something simple, one power cable and thats it. Thats exactly what laptops offer. When they do break its easy for them to just close the screen and carry to the nearest service depot.
/w 256megs of ram, only got a 6.5gig HD, but thats where the lan comes in.
I, on the other hand, just buy used laptops off ebay. No need for a 1Ghz+ machine for just running firefox and ssh. When something breaks I got a junker that I can grab part from. My current machine has been through hell and back and still runs great, I think I'm on my 3rd battery, I keep lucking in and getting pretty desent batteries with the used machines. Hell I can even run some old games on this machine. My current laptop is a dell latitude PII 400
I still have a powerful desktop machine that I use for gaming and lately mostly just a media player. The laptop is more pratical for day to day use. I'm use to the keyboard, more so than my desktop's keyboard.
I can see desktops in their current modular/user upgradeable form becoming more expensive than laptops in a few years. Custom built desktops will become a niche for power users and gamers. While adverage users go for iMac like all in one systems and laptops, and business goes towards thin clients. Don't get me wrong, this isn't going to happen over night.
Power users use way more bandwidth than your average no-clue person. They would ofcourse charge what they charge now for level 0 and the higher levels would start atleast double. Thats captialism there for you. No ISP wants power users unless they can find a way to profit from them.
In general your going to have to find books on each specfic technologies. Most of the general ISP resources are from the mid 90's when dial up was all the rage.
/w OC-3 cards, multilayer switches, DSLAM's and wireless bridges. Got a chance to mess around with a good mix of different layer 1 and layer 2 technologies along with a few different routing protocols. One course I totally underestimated was level 2 to 4 switching, I had no idea that ethernet switching was so complex.
An ISP sysadmin is a very specialized area and over the past few years the field has gotten smaller and smaller with everyone buying out everyone else. So your pretty much on your own.
I've just spent the past year doing Telecom Eng. Technology at the Collage of the North Atlantic. The courses included L2 to L4 switching, tcp/ip, network cabling, LAN's and DSL(Why they put these together), digital telephony, voip and prevoius to that I did a mess of eletronics and math courses. That broad range of courses still left out alot of topics which I covered on my own. I really made of the best of all the lab time I could get, built my own model internet out of old 2600 serise routers, a few linux boxes
Anyway my point is, you have to understand the underlying technologies and from there build a working knowledge of how to use your equipment at hand.
Oh yea and don't underestimate the power of linux in networking. A linux box can handle just about any task/service above layer 2, just gota get good hardware and with the money you save not buying cisco buy two or more, so you can always have a hot backup.