Things to Fix on the Web Part 1 – Downloads

Thought I’d start a quick little series of posts about things that bug me about the web today, and things that I’d love to see fixed and resolved in the next 5-10 years. Chances are these things will actually get worse and not better, but I figure I’d better start my wish list now.

First thing seems pretty simple… downloads.

The Problem
Remember back in the bygone days of the web when someone had something to download that wasn’t a link (ie: game, music, video, program, update, patch, etc) they’d create a link in their HTML page that looked something like this:

<A HREF=”/files/update-1234.exe”>Download update 1234 here</A>

You (the user) would click the link and lo and behold, the file would download.

It’s been ages since I’ve seen one of those on any “major” site.

What happens more often than not these days is something I bitched about in 2003, except it’s gotten worse I think. First of all, bandwidth costs money, so no one hosts their own files. Second of all, if a user just clicks a link and the download starts, there’s no opportunity to put ads in front of their eyeballs.

So these days the scenario goes something more like this… Click the link, go to a second site run by a third party which shows you the sites you can download the file from (ie: filefront, fileshack, etc). You click on one of these links and are taken to the file download site and their list of mirrors. East coast, west coast, listed by country, capacity, usage, etc. Clicking on one of those should give you the file right? Nope, now you’re on a page telling you that you can be cool and either login or register for your uber/gold/platinum/premium/monkeybum membership which lets you bypass this page and get files really fast, or you can continue on as an unwashed swine of a user, shunned and using the “free” servers.

Well, sometimes, other times you get one of the services which doesn’t allow you slacking freeloaders to download without paying, and if that’s the case you have to go back and start all over again.

So now you’ve selected the free servers you get your file right? Nope, now you more than likely get a page with a slow counter on it, forcing you to wait 30 seconds, 60 seconds, maybe a minute and a half to get your download. This isn’t them making you wait until a download “slot” is available most of the time, it’s simply to entice you that maybe the $9.99 a month would be worth it to pay to justfuckinggivememyfilenowdammit. Should you decide to wait though, when the download counter finishes you need to now (finally) hit the link that it gives you. If you get bored watching the ads (did I mention that every page so far has been splashing ads at your eyeballs as much as possible?) and switch to do something else, you might not come back to the page until your download has “timed out” and you have to start the whole stupid dance again.

Even worse, because of the glory and beauty of AJAX, javascript and similar webdesigner magic, the link you do finally get to click to get your file (remember we’re now 5 or 6 levels deep to download our game patch) isn’t a link at all, but some fancy JavaScript call which returns right away, but your browser just sites there for a while as remote servers pull themselves out of their slumber and think about actually sending a file in your direction. The ads being plastered at you work well though.

So that’s the problem, from the end user perspective of course. From the web designer point of view it’s even worse. You have to coordinate with another company or companies to host your files, deal with all the extra coding that’s not the simple “A HREF” to figure out where to send clicks on the ‘download patch’ link, etc.

Thing is, what to do about it?

The Solution
The thing that worries me about this particular niggle I have is that it’s a hard one to fix from the point of view of the internet model that users and advertisers have now.

  • Companies can’t afford to host files and pump out the terragigabits of speed and bandwidth needed to send their 1.8G patches out to thousands of users.
  • Companies are somehow making money from blasting ads at eyeballs
  • Companies are making money getting people to upgrade to the gold/platinum/premium download servers
  • Users are used to advertising and probably feel odd clicking a link to a file that just downloads (a virus! must be a virus!)
  • For users hard drive space and bandwidth are getting cheaper and cheaper

So from all of this I had a couple of ideas.

My original thought was that a new tag could be added to HTML, say something like <DOWNLOAD>. This would have configuration to allow the designer to give the user one link to click on, and that link would magically figure out how to get to the file, and the user gets the end experience of clicking on a link and a file starting to download, but on the backend server loads are checked, and the user request is automagically routed to a server in a pool in their area with the lowest use.

Great idea for the user if I do say so myself, but there’s so much advertising done now, and people are so used to getting things for free how could that satisfy the monitary needs of the publisher and ancillary companies they work with (distributor, game company, etc).

Some games now are almost completely self contained for updates. Systems like Steam are making it so that users don’t have to go anywhere to download patches or updates, they just sit down and know their game is up to date. There are others that do similar I’m sure as well.

Course, in some ways Steam isn’t much better than the old “guaranteed virus free” Active-X download programs from the days of yore, it’s got the upside of auto-patching, and being able to download demos and even buy new games right from within the interface, but still manages to toss lots of ads and more importantly, UI cruft at you (it helpfully pops up a window on start with “news” about new games for example).

Much as I’d like it, I don’t see online advertising getting any less online, and as with many things, making it easier for the user generally means making it less profitable for the companies.

I really hope that something does happen to make downloading files (and large files) something that “just works” for the users.

2 Comments on “Things to Fix on the Web Part 1 – Downloads”

  1. No need for the <download> tag — you can (and if you can’t provide enough bandwidth yourself should) put the file on a CDN (Content Distribution Network). The CDN will figure out all the load-balancy stuff to ensure there’s enough bandwidth for all your users, and they’re getting inexpensive enough with services like Amazon’s S3 that they’re not a huge pain in the arse to set up.
    The sticky bit is the advertising. It’s not so much the users that have got used to the presence of advertising that’s the problem, it’s that the publishers have built that advertising revenue into their system.
    It’s funny though, in that the rigmarole you describe is pretty much limited to game downloads. The worst I tend to experience is Sourceforge-style downloads, in which you have to pick/autoselect a mirror before getting access to the file (and there’s almost always a “direct link” available to click on).

  2. This is doubly annoying when you’re trying to get a URL to wget to a remote host rather than download it and re-upload it. Even sourceforge makes this difficult (although, they’ve got better lately, and you can hunt around for the direct download link).
    The problem isn’t a technical one though. The bandwidth issue is one thing, for sure, but the primary reason you’re seeing this is advertising. The ability to get multiple impressions in front of a captive audience is worth a lot of money, and when money comes into play, don’t expect it to change any time soon.
    I think we’re a dying breed – the web purists.. I don’t run ads on my sites, I don’t require anyone to pay for anything. It costs me money, but I do it because I love it, not because I’m trying to get rich.