Product Review: Websnake 1.23

In short: Great interface, poor engine. Wait for a new version, then run a trial version through a torture test before you buy.

The Web's a great thing, but one of its problems is that you can't take it with you. If you're not connected to it, it might as well not exist.

Offline Web browsers fix that. They grab an entire site, download it to your machine, then patch up any broken links. Now you can stick a whole site on your laptop PC and surf it while you're at the beach waiting for your plane.[1]

Websnake's strengths

Websnake 1.23 does precisely this and more, and less. I bought version 1.2 in hopes of getting something that worked, then upgraded to 1.23. Installation was a snap, and its interface was great. A wizard guides you through setting up a "snake", that is, picking a site to grab and setting some options. Nice and friendly. You can set more advanced options manually in a separate step.

The manual options are really powerful. I particularly like the fact that you can set several different criteria for ending a session: maximum total files downloaded exceeded, maximum HTML files exceeded, space left on disk below a certain level, total downloads greater than a user-defined limit, etc. You can exclude files based on file types (no MPEG or AVI for me, thanks). Etc. Nicely done. It lets you choose whether you want to grab pages from the original URL's location and below, from the same server, the same domain, or just follow every link. You can even schedule sessions to run later.

Websnake's fatal weaknesses

The problem comes when trying to actually grab a site. I can't figure out how it works sometimes. I set the maximum number of HTML files at 60, and at 23 files it cancels due to having exceeded the HTML file limit. Hunh? Does it count the number of references to pages, rather than the pages themselves? In the end, I have to set the limits artificially high to get what I want.

It also has problems resolving links in downloaded HTML. It gives you three options for links that point to things it didn't download: point to the original URL, link to a message file, or ignore. Sometimes this worked, sometimes it didn't. Also, if stopped in the middle of downloading a site, it sometimes doesn't fix links at all.

Websnake 1.23 has more problems.

A lot of these problems were around in version 1.2. Since that release, the product was sold to a different company. 1.23 seems a bit better, but still takes a lot of playing around with the options before I can get it to download the right files... and then I have to hope it fixes the links correctly.

Frankly, I would consider this software as an alpha in terms of quality. Maybe they'll come out with a new version that's actually worth $30. I won't hold my breath. It's a pity, since the interface is so nice that I want the darn thing to work for that reason. At this point, though, I'm willing to settle for less - so long as it works.

A point in their favor: the new owner's tech support responded to my questions within a day. After the go-round with the old one (including receiving a message saying that my last email had been in their queue so long I should resubmit it), that was heartening.

Take a look at the open source HTTrack, which does everything Websnake promised to do.


1. There are better things to do on a beach than look at Web pages.

Last updated 8 February 1999
All contents ©1999-2002 Mark L. Irons