On their blog today, web site monitoring service Pingdom took a look at web hosting services ten years ago and compared them to today's hosting services to see what has changed. The answer -- prices have gone down while included storage space and bandwidth have increased. Or, in other words, hosting hasn't changed much, but it has become a commodity service. Did many hosts miss a golden opportunity?
Probably the most interesting part of Pingdom's post was the comment it received from a reader named "Matt," who wrote that hosting has become a commodity service and web hosts have not done much to innovate in the space over the past ten years.
"How I read this comparison is that web hosts haven’t really done much in ten years but drive up some core specs. Most hosts are still hosting in the same manner and architecture that they did 10 years ago. Web hosting is a drive towards commodity. What really happened is that the web hosts, who could have been at the forefront of the hosted application space, squandered the decade offering the same old thing with little to no innovation." -- Matt
While Pingdom really only looked at data and bandwidth (and forgot to adjust prices for inflation), a quick visit to most shared hosts today confirms what Matt said. In general, hosts haven't changed much. Certainly there have been improvements in server architecture, control panels, and newer versions of scripting languages, libraries, and databases are running on faster servers, but in general, things are the same as they were 10 years ago at most shared hosts.
Matt is right that hosts have squandered an opportunity to innovate. I'm not exactly sure what he meant by "hosted applications" -- I don't think it would have made much sense for hosts to try to get into offering web-based consumer applications. That's a completely different business that is leagues away from hosting and it is doubtful that expanding into that space ever even crossed the minds of execs at most web hosting companies.
Hosts are, however, starting to innovate to create the environments necessary to power the new breed of web-based applications. The advent of cloud computing has resulted in hosting options that didn't exist 10 years ago. Amazon's Web Services stack, Rackspace's Mosso (which we just wrote about), Media Temple's Grid Service, and Joyent's Accelerators all offer new innovations in the web hosting space.
What do you think? Did web hosts let themselves stagnate for ten years and miss out on opportunities for innovation?
Photo credit: Ronnie Garcia
Listed below are links to blogs that reference this entry: How Has Web Hosting Changed in 10 Years? Not Very Much.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/3365
On their blog today, web site monitoring service Pingdom took a look at web hosting services ten years ago and compared them to today's hosting services to see what has changed. Read More
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
I'm a full time php developer
One big thing that has changed is thats VPS's (virtual servers) are now mainstream. A VPS gives you all the benefits of a dedicated server (you can install and config it however you like), lower performance though a far lower cost.
Also the stats reports that you get tend to be better (though google analytics is even better)
And yeah standard webhosting hasn't changed much, but it really doesn't need to. I'm perfectly happy working with a bare bones $5 a month account on a shared host running control panel. Setup database using phpmyadmin, upload .php files. Done. I really don't see the need for any additional features.
One big hole is the lack of shared ruby on rails hosting, but that's more to do with the fact that the founders of it put zero thought into running it on shared hosts rather then lack of innovation.
I'm guessing what's being implied here is that by offering different products, webhosts could have attracted non-developer crowds. That would be a sort of weird argument. There are dozens of 'canned-website' solutions being provided by lots of players i.e. Yahoo Small Business e-commerce solutions. It would be kinda dumb for webhosts who usually run with a skeleton crew to start doing software development.
Posted by: Steve Boyd | February 20, 2008 9:39 PMSteve makes a very good point - *should* webhosting have changed much? If it changes too much the company is no longer a web host, but in another business entirely. Witness companies that provide online storage - are they webhosts? Or software companies delivering the software as a service?
A decade ago a lot of people made the argument that corporations should outsource IT functions like email... but that ASP business turned out to be very much a mixed bag and, when companies pulled back from that a lot of ASPs died. Even those who didn't aren't really hosting companies.
If web hosts had changed radically would they still be web hosts?
Posted by: rick gregory | February 20, 2008 10:11 PMI guess I'm kinda surprises that "cloud computing" and more services/hosting similar to it haven't been developed and perfected further. Amazon really has a great thing going, but hasn't been dumbed down for the "average user" yet, not that an average user might need distributed processing and storage though.
From what I hear and understand, companies like Mosso and Media Temple are heading down the right path, but it hasn't been perfected yet. There's still not the flexibility of a dedicated server or VPS with them in terms of what software is running, but there's so much more flexibility in terms of processing, bandwidth and storage.
Posted by: Drew | February 20, 2008 10:17 PMWe need to understand that even if you use various new "Web whatever.0" things if you are an average website owner you wouldn't need more than a fraction of a commodity server. In most cases that would be a very small fraction. The long tail of average websites is really very, very long. So you don't need much platform innovation - changes in http/database server architecture, because you don't need to scale. The platform is improved incrementally by software and hardware vendors - Apache, Microsoft, Intel, RedHat, HP, IBM etc. The speed and reliability is improved mostly on that side. Also, it is easier to improve reliability on the host's side if the technologies have been used and tested for a long time - there are more technical personnel available having good command of those technologies, there is a lot of technical information etc. As one of the previous comments says VPS technology appeared not so long ago and is widespread now but it is just a way to improve the use of those same technologies - http/ftp/mail/database server, several scripting engines. Actually, with VPS you just get more control over those technologies.
New features that web hosting companies do provide and that do add some significant value are mostly services on the next layer above that. Ten years ago a full-featured web interface to the database server or a webmail were not the features a typical hosting company would provide. Now this is very widespread and maybe even is a must for many website owners searching for a proper hosting. There are lots of hosting packages with some popular CMS, blogging or forum engine pre-installed. This is I think the easiest way for web hosting companies to provide some added value to their consumers and I believe the changes in "an average hosting service package" will be mostly in that direction for maybe next 1-2 years.
Posted by: alibloomdido | February 20, 2008 11:35 PMSurprised there's no mention of Amazon EC2 here. It's most definitely a hosting product, even though Amazon mistakenly positions it as a developer product.
Posted by: Jeffrey McManus | February 21, 2008 9:09 AM@Jeffrey: There is mention of Amazon's entire web services stack, which includes EC2 (and S3, and SimpleDB, etc.). :)
Posted by: Josh CatoneAs someone who has suffered (along with many of his referred clients) from unending outages on the MediaTemple Grid Service, I assure you that its "innovation" has been far worse than my experiences on good ol' shared hosting.
Posted by: Jeremiah | February 23, 2008 9:05 AMFollowing on from a talk I did last night at the AWS User Group in Newcastle(albeit not about AWS!), I have to agree that there is a number of interesting companies who are finally innovating in the web hosting industry.
Amazon has certainly led the way in this respect, then there's Joyent, Mediatemple, GridLayer (and all the 3Tera customers), GoGrid, Mosso & last but hopefully not least, us with FlexiScale.
Having provided web hosting services for the last 11 years, we've truly seen the scale at which the use of the web has changed, both in quantity and complexity, and it really does need a new generation of technology to keep up.
In reference to point 4, I agree that virtualisation is the key that is enabling this next generation.
I'd like to think we are on the right track now, even if it's still evolving.
Posted by: Tony Lucas | March 12, 2008 2:30 AM