Month: September 2004

You know me, your friendly neighborhood license examiner…

When we last spoke I was all in a tizzy about the Movable Type License, well it’s time to move on, and look at another license. Next!

I signed up with Geocaching.com, you know the deal, create an account, log in, use the site, you do it everyday. Now, you can download what are known as ‘Waypoint files’ which have an extension of .loc and are really just very simple XML files. I’d show you here, but I don’t think I can, so I’ll describe it. (I hope that’s ok.) They contain information about a cache, an id, a name, coordinates, and a url that points to the Geocaching.com site. All good? Sure…

If you are not logged in to the site, it tells you that you have to log in to read the Waypoint License Agreement. Once you log in you can then read the Groundspeak License Agreement. I know, I just gave you two different names, which both seem to point to the same thing. It’s confusing like that… When I get to the page in question it tells me I’ve agreed to the license agreement, seemingly even before I’ve read it. It then presents a copy of the agreement for your records.

A copy for your records? I wonder if I’m the first user to actually make a copy of it for my records, let alone read the thing…

Oh, but I did read it, that’s what the friendly neighborhood license examiner does!

I don’t think I can excerpt any of it here, because I would probably violate it, but I’ll give you the gist of it.

First, they (“they” will be “GROUNDSPEAK, INC.” for the rest of this.) retain exclusive ownership of the data. I know, the users entered the data into the system, and perhaps they gave up their rights to it at that point, but let’s move on…

You can make one copy of the original data for archive purposes. Does this mean if you backup your hard drive more than once you are in violation of the agreement? I think so…

There’s plenty of nice bits about what you can do with the data for internal purposes, the data can pretty much be “pushed, filed, stamped, indexed, briefed, debriefed, or numbered” as long as you do it internally. That’s good.

You cannot sell, rent, or lease the data. This is reasonable, and for a commercial entity, heck, for any entity, this is the “right thing to do” and I totally agree.

But… you also cannot provide access to the data to unlicensed third parties. Does this mean if I upload data to Geocaching.com it then becomes their property and I can’t tell you about it unless you have also agreed to the license? Can I post it on my weblog?

Ok, this next one is just wacky. It reeks of some lawyer who has never used a computer. Ready? You shall not reverse engineer or disassemble the data format, because, you know, you might try to dupe their proprietary and copyright-protected data model and export format. This is an 8 line XML file, which you can read with almost any computer manufactured in the last 20 years. If there is a job out there reverse-engineering tiny XML files, I want it! Seriously, anyone who has ever seen HTML can probably reverse-engineer this thing by quickly glancing at it.

Oh, there also seem to be a number of applications that can use these files. I wonder if all of the developers have agreed to some sort of license, or had to pay some fee. Surely they didn’t just reverse-engineer the format!

There’s one more bit worth mentioning. Something about how they will be willing to license the data to you for a fee. Yes, they will sell, rent, or lease the data to you for a price.

So what’s the deal? Well, Geocaching.com (dba Groundspeak) has to protect themselves, and their data, and have to make sure they can make money from it, and prevent others from making money from it, since that could cause them to not make money. That all makes sense right? Sure some things bother me, like my interpretation that your data becomes their data and you lose the rights to it. I had thought that when you create something, you are by default the copyright holder of it, is there a transference of copyright in this case? Maybe that’s in an agreement somewhere I missed. Probably when I created an account on the site. I’ll probably get kicked off soon and have to wander the wastelands unable to geocache…

Perhaps a Creative Commons license could be of use here, allowing non-commercial use of the data. Of course I always wonder about that too, what if I provided some web site with the data, free of charge (non-commercial) but then had ads appear on the site (commercial) how is that handled? I’d like to use the data to do interesting things, and I can – internally – but it seems that if I show you that data, I’m in trouble.

So why in the heck do I bring this stuff up? (Besides being a troublemaker who likes to complain a lot…) Because people need to be aware of what they are doing with their data. We’ve talked before about exit strategies and the importance of such in relation to your data, and we’ve talked about people agreeing to licenses and violating them. Exit strategy good. Data lock-in bad. Violating licenses bad. Reading licenses good.

Hmmm, perhaps I should have titled this “How to make money on open-source software”

Nonetheless, I’ve seem relatively few projects use ransoms for development. David Raynes mentions this idea in combination with Dropcash in an entry titled Using Dropcash for Feature Ransoms? Seems he hit the nail on the head, as Chad Everett tried it with his MT-Notifier ransom and David did one with his MT Dropcash Plugin Ransom that came out “more than successful” in his words…

I’m probably not an expert in the field, but over the years I’ve seen people do a few things, release their software as shareware, or “freeware”, or commercial, or open-source… and occasionally the open-source stuff will have a donation model, where you donate cash (usually via PayPal) or you send a gift (usually via Amazon) and in these cases you do the work (probably for the love of doing the work) and if you get compensated, well, that’s just a bonus. There’s nothing wrong with this model, and I hope it continues. Luckily open-source developers have things like Sourceforge to help with the infrastructure, bandwidth, and hosting costs associated with releasing open-source software.

Here’s how I see the ransom model working for feature requests and enhancements. You’ve released some application, but it doesn’t do X. Users request feature X, and in reviewing it, you determine that it’ll take Y hours to do, and you’d like to be compensated with Z dollars. You just need to figure out the X, Y, and Z part of it. Any problems with this? This should not violate any open-source licenses, as you’d still release the code once it’s complete, then anyone can have it. Is it fair to the people who paid that everyone else gets it for free? Yes, the people who paid really wanted that feature, and were willing to pay for it…

This is actually done by some companies that deal in open-source and/or custom development as well. There’s a gated community of users, maybe they’ve got 20 clients in the steel industry who use their product. If one customer wants a feature added, but can’t afford it, they may see if other companies want that feature as well, and distribute the cost. (Hmmm, perhaps this should be called “distributed cost development” or something.)

Now, for new software, it’s a bit harder. Someone who has never released anything isn’t likely to get any ransom money for an unreleased application, especially with no prior bits of code out there. So, if you plan to ransom and release something, you’ll most likely need to have released something of value previously. For instance, Developer Joe has this great new idea for Application X, and he’s already had 100,000 downloads of his previously released Application Y, and users love it. Based off of Developer Joe’s reputation as a good programmer, he might be able to get people to pitch in and pay the ransom for Application X.

Anyway, I think the idea is fascinating, and there are probably many ways to make money on ope-source software, and even commercial software that have not been fully explored yet.

I use Thunderbird to get mail via IMAP to the Exchange server. Now, Exchange is a weird beast, and I am not a Windows guy, so there’s a problem. Outlook lets you create rules (aka filters) that can run on the server or the client. I set up rules on the server using Outlook originally, and figured it was easier to do it on the server since I use Thunderbird in at least 3 or 4 places, and didn’t feel like making filters for each copy of Thunderbird.

So in typical Microsoft fashion, things don’t work. As I added more rules via Outlook, they were not working. I finally mention this to an ex-Windows guy, and he tells me there’s some limit to how many filters can be created on the server. Argh! You think Outlook would tell you this! But no… and it’s not like I’ve got hundreds of rules, it’s really not that many…

That’s ok, we can work around that. Filters in Thunderbird are fairly easy to create, though I’d still like to see a text file I can edit, like the the mailViews.dat file does for views. (Any idea where the filters are stored?)

As long as we’re talking about views, those things in the little drop-down menu, they are easy to create, and if you want to copy from one machine to another, find the file named ‘mailViews.dat’ in your profile folder. You can even reorder them so they appear in the filter menu the way you like.

Oh, the format is pretty simple, I would think it would be easy to share views quite easily:

name="Has Attachments"
enabled="no"
type="1"
condition="AND (has attachment status,is,true)"

A far cry from mork indeed!

Ok, you want one more file to mess with? Then try ‘persdict.dat’ which holds the words you’ve added to the dictionary when doing spell checks. You know, words like “diff”, “grep”, and “wiki” – ok, that’s it for today’s Thunderbird hackery…

I’ve got something I call vcalxical.pl, which is a perl script that logs into an Exchange server running IMAP, and then reads a mailbox that is really a calendar, and pulls out the VCALENDAR parts, and formats them into an ics-type file. You can import this into Apple’s iCal or into Sunbird.

Now, it’s far from perfect, but it works for me. Your Windows guys would have to have IMAP enabled, and I’m guessing it doesn’t take much to break things if they really want to. (For instance, at some point in time things changed from plain text to HTML, who knows!?)

The idea is this, when stuff gets scheduled in Outlook/Exchange (which you don’t use cuz you don’t use Windows) it’ll show up in iCal or Sunbird after the script runs via a cron job. Theory, anyway… I think it would be better to have the calendar on available via webdav, as I don’t know how the apps like having the file changed on disk willy-nilly. That way it really is a remote calendar.

I’ve been using it on and off, but not testing it too heavily. iCal seems to be popping up alarms to stuff. Sunbird, I’m not sure yet…

Still, we hack on…