We have some fairly large files that we need to make available to download from our website, and are trying to think of ways that we can make this a less painful process for them. Zipping the files is the obvious solution, but I am wondering whether it is safe to assume that all our users have access to an unzip type application?
We have a lot of files in PDF format, and that causes enough problems of its own, so we’d rather not add another layer of complication/confusion…
A link to the “free trial download” page for the necessary application (an unzipping application, in this case) is how it’s usually done. I would also include a link to Acrobat Reader so people can read your PDFs.
You’d be surprised how many people are incapable of downloading and installing an app like Reader. I handle the webmaster@xxx emails and most of my time is spent explaining how pdf documents need an application in order to be opened and how they can get it…
I would put them up there in zip format and provide a link to winzip.com; self-extracting .exe files would work, but some people are understandably twitchy about running downloaded executables (some companies actually have policies forbidding it).
You can link to Winzip’s website if you want, but it would be advisable to also state that Winzip is NOT free. It can be downloaded and evaluated for 30 days before it must be purchased or uninstalled. If your users are comopany employees perhaps the company can buy a site license and have it installed on as many PCs as they need. I’m also not sure if the self-extracting archives it creates can be executed in a 16-bit OS.
You might check out Download.com’s File Compression category and toy around with some of the free unzippers out there and see if there’s one to your (and your users) liking.
What’s happening is you are clicking on the link to the .exe file and the executable is actually running/executing on the web server instead of offering it as a download to your computer.
Don’t use self-extracting zip files (.exe) unless you know that any and all Macintosh users are using versions of Stuffit Expander that are late enough and sophisticated enough to reach past the EXE layer and grab the ZIP portion and expand it.
I would assume that if you have Windows 3.x users out there, you probably also have Macintosh System 7 users (conceivably even System 6 users). If you really want to accomodate your Mac users, give them the option of downloading Stuffit (SIT) archives instead of ZIPs, as really old versions of Stuffit Expander may not be able to deal with Zip files.
My main recommendation is to only compress files that you know can be compressed. If you’re making packages of large numbers of files this isn’t workable, but if you’re compressing single files its good to think about. For example, PDF files SHOULD BE precompressed, so zipping them will give negligable benefits.
Do not ever consider self-extracting executables. This makes your problem worse, not better. A. People using a platform other than the executable’s native will be confused. (Even though a standard unzipper knows to jump over the self-extracting code.) B. People using the platform aren’t going to trust and run your executable if they are at all smart. It’s a lose-lose situation for you.
It is never a wise business decision to exclude even 1% of your potential customers if there is no practical reason to do so.
Unzip utilities are available for all common platforms, even pre-windows DOSes, handhelds, etc. Rather than link to limited platform unzip programs, link to Infozip instead. There, they can find an unzipper for their platform. (It’s free, compatible, source code available, etc.)
Remember, the user is entitled to make any and all such decisions. Not you. You goal is to make their life simpler, not the other way around.
Other things to ponder: If you are concerned about the slow times of dial-up downloaders, don’t. Their modems (if modern and set correctly) will automatically compress/decompress data as it goes over the phone line. (Which is wher those magic “up to” speeds come from.) Adding another layer of compression on top of this hurts, rather than helps.
There are pdf2ps utilities out there so people can convert to Postscript if they don’t have a pdf viewer. (Ghostscript makes Postscript ubiquitous across platforms.)
There are a lot more things to consider if you wish to do things intelligently. Perhaps someone who knows these things should be hired.
Is it impractical to convert the documents to HTML? Most modern applications can write HTML, and it should be more universal than PDF. PDF is of course much better for formatting, but for most documents it’s not critical. I usually avoid PDF documents on the web mainly because of the 15 seconds it takes for the Reader to start up.
The other day I came across a 200kb zip file which expanded to a 220kb PDF file, which turned out to contain one page of text with some fancy background (watermark). It didn’t even have complex formatting, just a plain list. Don’t do that, it’ll make your customers mad. And waste disk space and bandwidth on your server.
Thanks everyone for the comments and suggestions - I am losing confidence in the concept of self-extracting zip files, especially after reading ftg’s post…
The files are very large Word files which (in an effort to be cross-platform) we have stored in RTF format - they compress very nicely, but I was trying to avoid having to force ludite users to install another application. Seems there is no way around it…