Google can take the web, yawn readers
World domination cancelled due to lack of interest
Letters Bad news for Google - it won't be taking over the world with its Web Accelerator any time soon. That's because the web is much less important, compared to other internet services, than the giant web search engines like to think.
"If they really want to sink their brand into this," yawns one mightily-unimpressed Reg reader, "Google will end up controlling a high end phone book and homework library."
That's one of the reactions to our take on the Cringely view of Google's Web Accelerator as a world-domination ploy, we discussed here. And similarly, with their focus on chat and games, the Asians might actually be on to something.
Pipex founder Peter Dawe has also penned a view of what went wrong with the Internet, which is a don't-miss. On with the show.
Google can take the web, it doesn't overly affect me and a large number of people I know who use the internet in more of the Asian way. We heavily use our internet connections for other items like games, irc, voip, web radio and TV
Another thing that we are actually doing is vpn'ing our home networks together and even developing our home networks so that we have central music and video stores and stream them across to our other machines dotted about the house. We're even playing with creating our own TV channels based on our large collections of movie files and VLC so that we can simply enter the streaming address and pick up on the stream from one of our servers and watch tv as we like instead of being restricted to programming schedules forced upon us by the TV channels - and using our PVR boxes we can add new content all the time to schedule at a later date.
It won't matter in the end. Our concern with the 'bare metal' of the internet mimics our earlier fascination with chip architectures, instruction sets and motherboards. These are now commodities and the average punter doesn't care how it works or who supplies the chips (neither do I for that matter).
The internet is infrastructure and in the future will be about as exciting and newsworthy as a root canal. General purpose home computers will gather dust as special purpose machines with cool designs and unbreakable software take over. Entertainment will bypass your computer and go directly to a home entertainment concentrator, IM will go direct to your mobile and news will be delivered direct to (ironically) your telly.
If they really want to sink their brand into this, Google will end up controlling a high end phone book and homework library.
Try to imagine how little I care.
As for the AsiaNet, with Google's recent crossover into China, we should watch for what sort of shrewd ideas they come up with over there. Online Games and Instant Messaging are infants (at least in terms of widespread useage) compared to web content and there's literally hundreds of possibilities for them in the future. If Google plays it smart and plays nice with the Chinese Government, they might have a lot of influence on how computer use in China evolves. Look at the effect they've already had everywhere else. Still, the thought of the entire internet being controlled by Google is absurd.
Even if Google somehow managed to find a way to blanket themselves over the entirety of the Internet, they would not kill it. We need only to look back for the proof. People still listen to Vinyl over CDs. People still use IRC over MSN Messenger. People still play MUDs instead of Everquest.
People will always use the Internet.
Hi Andrew, While there is a lot of uses for the internet and most consumers are out there trying them all (I wonder how many of them send away for the “Enlarge your penis with natural herbs” products), I think that there will be a point when the West starts to treat the internet like a tool again, rather than a toy and then you will see people start stopping to rely on the web so much. It is interesting that you mentioned the web replacements for non-web activities.
One of the things you didn’t mention was the way that IM services are trying to replace IRC as a means of direct communication (MSN and Yahoo! Chat rooms from within the IM clients as opposed to mIRC). I had a discussion the other day with a friend who said that she doesn’t go into chat rooms anymore. I was talking about IRC and assumed that she meant the same. But her idea of IRC was MSN Comic chat and the associated chat rooms. I didn’t have the heart to tell her that was not what I meant by IRC, I let her live in her own little world on that one.
These days, I do a little web surfing to maybe about 15 sites and I use the web for work purposes (I work in web development). Most of the time when I am on the web, I am either working or playing games or chatting on the IRC. Thanks to the glut of spam and advertising in emails these days, I tend to pick up the phone and call someone and re-awaken my social skills by actually talking to a person rather than a name on the web.
One day, I would like the web to be back to what it once was a decade ago. A pristine place where people come to do research and communicate with like minded individuals and not be plagued by popup’s popunders and self-installing XXX search toolbars and a million other useless things that unfortunately me and my colleagues in the web dev business has created.
Bob Cringeworthy is a widely-written journo, but let's face it, he's also off on his own planet somewhere. Google offers an "intranet crawler" hardware device offering Google's search capabilities over a customer's intranet. It's never made any headlines; any Google ISP, Google PC etc, seem doomed to the same fate. Personally, I really liked the idea of a 1U Google-in-a-box intranet search engine, and expected it to be huge for Google. But it wasn't. I suspect that you're giving Google credit for more ability than they actually have. What they do have is an enviable index of the web (and usenet, FWIW commercially). They've also got some good skills in combining information, like their adverts on Google Maps.
Their ability to run an ISP, etc, is a totally different matter. That's not how their network is built. Their network is built for read-only international queries, not localised data access.
Was there ever an 'Internet' to take over?
Well, yes, and no. I can't get this thought out of my head: Consider all the organisations with premises connected by VPN tunnels. They make closed networks out of the public network and will not go away because the alternatives are uneconomical - they certainly can't be taken over by Google. VOIP is another public network application that will change closed network telephony forever in exactly the same way - various people have speculated that Google might have a go here (dark fibre acquisition, anyone?), but there are a lot of vested interests and they might find it hard.
So, with just these two examples of closed networks operating over the Internet and the standard western view of web and email, we can envisage what I will call Schroedinger's Network if you like... it's both public and private at the same time - and there is no way to take over without making it entirely one or the other. There must be countless applications and protocols that we can only speculate about, all of them adding up to a greater economic whole than that of surfing or chatting on the truly public Internet, even if the data volumes suggest otherwise.
VPN's aside, I can think of many applications I have been involved with that used the public Internet for closed or proprietary uses - most of these were in the days of modems and were often embedded systems that uploaded data periodically - and yet only a very small number that have used truly closed networks (Teletext being one in particular that you didn't mention).
Not convinced? Close all the VPN tunnels or shut down Google... which would truly change the world? Google is just a niche player, albeit a very big public one, it would seem.
A few flames for us for taking Cringely's fascinating theory seriously.
You reported Cringely as saying, "Part of that has to come from Google assuming a larger role over time, taking responsibility for rendering Flash, for example" Normally I have a lot of respect for Cringely but this comment is cringeworthy.
Flash is a bit-optimized format for the rapid transmission of animated graphical data over the internet. Its compression ratio is millions to one over prerendered data. The only technically feasible way to do this would be to render Flash, then recode it into, say, MPEG-4. In real time. That's a lot more power than it takes to serve a page from a proxy. For every user at a Flash site. As they get more and more common. Moreso, each user would require far more bandwidth, and for users with caps (bandwidth is *not* free) viewing one 100K Flash file as a prerendered MPEG-4 is likely to take up his entire monthly quota.
There's a big reason intelligence doesn't go into the network, and it's not a matter of principle. It's a matter of bandwidth. While Google can save bandwidth it's offering a service that's useful. Nothing else makes sense.
You know Andrew, the more I read Cringely, the more convinced I am that he's a troll.
Even if he were just wrong-headed and ignorant, he couldn't be as consistently wrong as he manages. "If Google adds power to its part of the Accelerator, you don't have to add power to your end, meaning your old PC can last longer."
What rubbish. Web browsing is already just about the lowest-impact activity you can perform on a desktop, and the limiting factor is rarely the desktop hardware and almost invariably the connection to the net itself. Any upgrade by Google is only going to affect the net connection - delivering more data faster is, in this model, going to make old PCs wear out *faster* - or would if the whole scenario wasn't bollocks in the first place.
But wait! Cringely has an ace up his sleeve! Google will make your computer last longer by easing the CPU load by... "taking responsibility for rendering Flash."
What? Does Cringley actually understand how the internet works, or is he like your maiden aunt who thinks adding half a gig of ram will make the Food Network network load faster? Flash is a compressed format designed for fast transfer over the internet. Google web accelerator is a caching technology designed to speed up the transfer of files over the internet. That is, both of them are designed to make the most of limited bandwidth. Pre-rendering the Flash server-side would only massively increase the time it takes to deliver it, which is the *exact opposite* of what the Accelerator is intended for. Great logic, chuckles.
Not to mention that displaying full-motion video is going to be scarcely less intensive than rendering Flash in the first place. Unless Google is so clever that it can bypass entropy and deliver smaller files that take less power to decode and display the same data?
So I can disagree with his logic and his conclusion without affecting the issue; can I disagree with the entire premise the article is based on without affecting it? My guess is no - which rather begs the question of why The Register is prepared to do so.
And now for where it all went wrong. Take it away, Mr Dawe.
As the founder of PIPEX the first UK commercial ISP nearly 20 years ago, I feel I may have an interesting perspective on corrosion of the Internet.
The first problem, was the introduction of Network Address Translation, with dynamic addresses, the ability to shut down any malign agent was forfeited. further NAT required the use of port mapping, this has lead to many services being served by one port number, rather than a service per port. Thus, when I open up a port for Instant messaging, I introduce a ftp service as well.
NAT also destroyed the peer to peer nature of the network, Us early pioneers understood that access permissions were essential, if control was lost on permissions, all is lost. . Port mapping and loss of permission control are, like most pollutants impossible to take out of the system. Even if we re-implemented using, for example IP V6, The port mapping code would remain in the server software.
Bearing in mind the ubiquity of 'bad application code' implementing a closed network does not help the same applications get run. Minitel and SMS are not analogous to a private network, they are narrow applications which though they are less attractive to pollute, they are still polluted. I'm sure I'm not the only parent to have a son 'mugged' by premium rate SMS services or receive SMS Spam inviting me to send an SMS to an unknown admirer.
Don't look for encryption to completely rescue us either. too many people are ignorant, lazy, stupid or criminal to maintain a high enough level to maintain probity. Look at the number of successful phish frauds
What hope? Well if there was ...
1) A cut down web browser that can ONLY display text, pictures and play audio and video. No Java, No nothing ( I'll put up with not being able to access sites that require me to strip naked and bend over! no matter how attractive the offer)
2) An IPv6 internet where address and port mapping are outlawed
3) All communications digitally signed, and emailers and browsers that automatically check
4) Signed applications, and an operating system that automatically checks.
5) A digital identity service that allows the cross checking of signed transmissions . But none of these are going to happen, so its back the handshakes, pen, post and cash.
Mr Peter Dawe CEO P Dawe Consulting Ltd
One web for the Californians - many nets for the rest of us? Sounds good. ®
Sponsored: The Nuts and Bolts of Ransomware in 2016