Author Topic: 7 Reasons Not to Direct Link to a Developer's Download Files (Read 43,666 times)

mouser · « **on:** March 17, 2009, 04:50 AM »

DC member April Russo (app103) has an essay in her blog today that is worth reading if you run a website and are guilty of this sin..

Quote

As a developer, I can't help but be more than a little bit bothered by the fact that some bloggers chose to direct link to application download files (.zip, .exe, etc) in some of the articles they write, rather than the page on the developer's site, in which the download links can be found. If this is how you do things when you write articles & reviews, then I'd like to inform you that this isn't a very nice thing to do, for the following reasons:
...

http://cranialsoup.blogspot.com/2009/03/7-reasons-not-to-direct-link-to.html

Deozaan · « **Reply #1 on:** March 17, 2009, 05:07 AM »

Very good article! All points made are extremely valid.

app103 · « **Reply #2 on:** March 17, 2009, 05:07 AM »

While I was happy to see them mentioned, I was rather upset when I saw that 2 of the 3 DC applications mentioned today on makeuseof's blog had improper linking, one of them belonging to skrommel, the other being mine.

Their staff writers should know better than to do stuff like that. They are supposed to be experienced, knowledgeable, professional bloggers.

The comment I left on their article didn't appear for some reason (probably caught in their spam trap), so I reworded it a little and expanded it to turn it into a post on my own blog.

Consider it a vital lesson in your "Blogging 101" course.

rgdot · « **Reply #3 on:** March 17, 2009, 05:13 AM »

Nice article, never done it myself. Either to the front page or to the program's details page, I would say one shouldn't even link to the download page itself if possible.

app103 · « **Reply #4 on:** March 17, 2009, 05:15 AM »

Quote from: rgdot on March 17, 2009, 05:13 AM

Nice article, never done it myself. Either to the front page or to the program's details page, I would say one shouldn't even link to the download page itself if possible.

On a lot of sites, the info page has the download link.

mouser · « **Reply #5 on:** March 17, 2009, 05:16 AM »

App i do think you missed one HUGE reason not to direct link, which is that the download locations for files frequently changes. When you direct link to a file, that link will be totally unusable if the author ever changes file name or location.

And let's not forget that almost everything you say also carries for sites that host a local copy of the download on their own site and link to that (although i don't think it's a bad idea to link to a backup mirror copy in addition to the program website in case the original download or site dissapears).

rgdot · « **Reply #6 on:** March 17, 2009, 05:18 AM »

Quote from: app103 on March 17, 2009, 05:15 AM

On a lot of sites, the info page has the download link.

True, you are right

app103 · « **Reply #7 on:** March 17, 2009, 05:31 AM »

Quote from: mouser on March 17, 2009, 05:16 AM

App i do think you missed one HUGE reason not to direct link, which is that the download locations for files frequently changes. When you direct link to a file, that link will be totally unusable if the author ever changes file name or location.

You are right. That's so obvious that I didn't think to put it in my article.

Quote

And let's not forget that almost everything you say also carries for sites that host a local copy of the download on their own site and link to that (although i don't think it's a bad idea to link to a backup mirror copy in addition to the program website in case the original download or site dissapears).

In most cases you technically aren't allowed to do that, since it is the author's legal right to have full control over distribution of their intellectual property. It can be a copyright violation to host a mirror of anyone's files without the author's permission first, even if the author's site disappears.

Technically, without permission, you have to wait till the copyright expires and it enters the public domain, which won't occur until many years after the copyright holder's death. Currently, there is no computer software in the public domain unless the author has specifically and officially placed it there himself. Even hosting abandonware can be considered a copyright violation. This includes old C64 software, too. It's still protected by copyright.

Deozaan · « **Reply #8 on:** March 17, 2009, 05:34 AM »

EDIT: Removed content in which Deozaan forgets his manners and says really rude things which he is now ashamed of.

mouser · « **Reply #9 on:** March 17, 2009, 05:38 AM »

deo that is mean and unfair not to mention a downright rude thing to say to someone who took the time to write about software on this site

one of the nicest things about DC is that we try hard to be constructive and help people learn to do things better. as a long-time dc member you're better than that.

when people make mistakes let's try to find constructive ways to point them out and show them what's wrong and how to fix the problem.

Deozaan · « **Reply #10 on:** March 17, 2009, 05:47 AM »

Quote from: mouser on March 17, 2009, 05:38 AM

deo that is mean and unfair not to mention a downright rude thing to say to someone who took the time to write about software on this site

You're right mouser. I don't know what got into me.

I'm sorry to everyone, especially to Damien.

app103 · « **Reply #11 on:** March 17, 2009, 05:52 AM »

Quote from: mouser on March 17, 2009, 05:38 AM

when people make mistakes let's try to find constructive ways to point them out and show them what's wrong and how to fix the problem.

I have done this too many times to count, and I'll keep doing it whenever I see it.

Here is another example of educating people not to do this, and why. I came across an article about a tray utility that the writer linked to the application file instead of the developer's page for it, and had I not bothered to go to the main site & look for the app, I would have ended up with the wrong version for my OS.

Unfortunately, some people just don't learn, or they don't care enough to change their links, since it's been 6 months since I commented and he still hasn't fixed it.

PhilB66 · « **Reply #12 on:** March 17, 2009, 05:54 AM »

How about both a direct download link and a link to the developer's/author's Website.

mouser · « **Reply #13 on:** March 17, 2009, 05:54 AM »

Quote

You're right mouser. I don't know what got into me. I'm sorry to everyone, especially to Damien.

no worries -- we all have those moments

lanux128 · « **Reply #14 on:** March 17, 2009, 05:55 AM »

this is an increasingly difficult situation to handle. as an example here's a dude who runs a blogs and provides support services but can't link properly to a website.

app103 · « **Reply #15 on:** March 17, 2009, 05:58 AM »

Quote from: PhilB66 on March 17, 2009, 05:54 AM

How about both a direct download link and a link to the developer's/author's Website.

No, because most readers will just click the download link and it will have the same results, with regards to reasons 2 & 3.

justice · « **Reply #16 on:** March 17, 2009, 12:09 PM »

Shouldn't be too hard to check the referer and direct to the download page if it's a direct link? Not sure how effective this system is though. Also the developer can create a startdownload page that meta refreshes to the download link - that way most people that copy the link won't copy the actual file but the page that starts it.

It's all about keeping control really for what is their responsibility. The reason I think most people direct link is because its more convenient then clicking 'download' then having to go through another webpage to find the 'actual download'. Pointing to a startdownload page however would overcome this.

tinjaw · « **Reply #17 on:** March 17, 2009, 12:32 PM »

Justice said what I wanted to say. That is how I would handle things. Don't even have a direct link on your site. Have the link be to a script (CGI, PHP, whatever) that only serves up the file if clicked from your page.

app103 · « **Reply #18 on:** March 17, 2009, 03:19 PM »

It gets really complicated when you have download sites that are allowed to direct link because they are pulling in data from your PAD file. You can't really add just the download sites to a whitelist because in many cases you don't know all of them if you have your PAD file submitted to a PAD directory. Plus, even if you could whitelist them, that list could change daily.

The alternative is that you could do what you said for all other downloads, and have a separate file location just for those sites using your PAD file, but that could easily double the disk space needed to host your site if you end up needing to use 2 different copies of the file and that might not be possible for an author that has cheap or free hosting with a small disk space allowance or his web host only allows static html pages. (no cgi, php, or whatever)

Also, meta refresh may not work if the user has that disabled in his browser, and many do. Almost everyone I know on dialup has it disabled because it gets quite annoying on sites where you are trying to load a page and it keeps refreshing before the page is even halfway loaded.

app103 · « **Reply #19 on:** March 17, 2009, 03:38 PM »

My comment was finally posted on makeuseof, and received an official reply:

mouser · « **Reply #20 on:** March 17, 2009, 04:11 PM »

Great

Makeuseof is a good site

Nod5 · « **Reply #21 on:** March 17, 2009, 05:17 PM »

I have mixed feelings about this (speaking generally and not about the case at hand). I acknowledge that app103's arguments have some force. But then there's the convenience factor. I really like repositories-ish ways to download new versions of programs. Like http://www.filehippo.com/updatechecker/ . Such services can help people keep their software up to date a lot better. That has security benefits for the individual and for all of us since vulnerable computers leads to larger botnets that threaten us all. Maybe a middle way is possible here? Blogs that do separate posts about software mainly aimed at getting readers to know about a software for the first time (makeuseof, lifehacker and so on) should do as App103 suggests. Other sites like filehippo can host/direct link the files. But maybe they could also add some reminder or plea on the site about the value of visiting the source site, make a donation, help out in the official forum and so on?

edit: spelling

rgdot · « **Reply #22 on:** March 17, 2009, 05:59 PM »

The recognition to the software author should at least occur once, if a service like filehippo provides updates it must make arrangements with the software author to do so for things such as security updates, if for example the user doesn't want to use a built-in 'call home check for update' feature, going from v2 to v3 (ie a non-security update) straight from an updater may not be a good idea anyway because of changing requirements. Of course it's nice to have a central place to hear about new releases whether they be lifehacker or filehippo but updating straight from there is a bit different IMO.

mouser · « **Reply #23 on:** March 17, 2009, 06:38 PM »

Nod5, i think you've hit the nail on the head regarding sites and tools which need to direct link to a download. The most important thing as far as i'm concerned is that such services include a clear link to the original program web page.

Darwin · « **Reply #24 on:** March 17, 2009, 07:19 PM »

Just a thought - if I like an application that I download enough, I usually try to keep it up to date and to visit the author's website to see if s/he has any other applications of interest and to find out a bit about them. Thus, even if I download an application via a blog that linked directly to the download, I'd wind up on the author's site. Actually, come to think of it, if I was interested enough in the description I'd more than likely to visit the app's website before installing it because I'm pretty paranoid about what goes on my machine.

My overall point is that I suspect that there are those people who will not donate/visit an application's website no matter what and those that will, even if they have to go through an application's installation directory to pull a webiste link out of a readme.txt file.

The above is not to take away in any way from app's arguments, just a general observation about human nature.