jeff.stevens wrote:Is the attempt here to achieve a standardised format that interested individuals could then implement on their own platforms for personal use, or to come up with an application and repository hosted on the Internet where anyone could submit an observation?
Well, "none of the above" and "all of the above" and some places inbetween. As I said in my initial post, this started because I was making "backup" copies of my logs onto my website but wasn't happy with the "quick hack" approach that I took. The logs are written directly as HTML and, as such, the data is obscured by the markup language. My thinking was (and it's an obvious approach) that it would make more sense to write the logs with some sort of markup that emphasises the meaning of the data and then work from that. Hence the reason that I went looking for XML schemas for doing this sort of thing because translating them to HTML and other formats is easy enough.
So, having decided to do this, I then got wondering about if I should just invent a markup that fits my own style of log keeping or if someone has already worked on this and produced something. That then, of course, got me thinking about if anyone like the SPA or the BAA (I've sent an email to the BAA about this) has looked into this and have come to any conclusions.
So, it was more a case of "I'm going to do something about this anyway but if there's an emerging standard out there it might make sense to use it or a subset of it".
jeff.stevens wrote:It will be interesting to see if your research turns up any more sites or projects.
I've not found any other projects so far.
jeff.stevens wrote:I put together a small application for personal use a few years back (non-web based), but found that I actually preferred using a hand-written log.
Actually, the application side of things was the last thing I was thinking about; I was thinking more about the format of the data. I do now have a little project underway and, so far, I'm writing the XML data by hand -- I might look to writing an application for creating/editing logs at some point in the future but that's a secondary concern.
As for keeping logs in a notebook: as I said in my first post, that's what I do now. I'll also carry on doing that. I'm not looking into this with a view to replacing my notebook, it will always be the method via which I make my notes "in the field". While I do generally carry a Pocket PC with me everywhere I go and while I do own a laptop too I wouldn't dream of using them to make notes "in the field" -- for many reasons. There are a few reasons for me keeping an electronic version of my logs
and the main one is that I've got a backup copy should anything happen to my paper notebook. On top of this it's easier to index the electronic version (currently I maintain a general index
and an index by messier object
(the latter being possible evidence of the slowest Messier Marathon ever
jeff.stevens wrote:It's true that it is of no relevance to anyone else, and the observations are of no scientific importance, but I get great enjoyment from picking it up off the bookshelf and looking back through it occasionally.
Hmm, I beg to differ. I'd be the first to say that my current observing sessions have no scientific importance at all and I'll take your word in regard to your logs, but that sort of misses the point. Other people's logs serve a useful purpose other than "scientific importance". For example: it's very useful to read other people's logs and see what sort of results they're able to get with what sort of equipment and under what conditions. So, if I'm seeing whatever
feature (or lack of) on Jupiter and I'm not sure if I should/shouldn't be able to do it with my setup it's very helpful to read the logs of someone with a similar setup and know if my experience is the norm or not.
jeff.stevens wrote:However, if I knew that my contributions were to part of a wider knowledge base, then this might motivate me to use an electronic log. Coming up with a standard template is always going to be tricky. Would it be a generic one, or a more detailed set of templates dependent upon the type of observation being recorded, e.g. data relating to an occultation event would differ from that of a meteor watch or a planetary observation? I'm a great believer in keeping things simple, the more complicated something is, the less likely it is to be used.
Well, the two projects I've looked at so far seem to be about making it as detailed as possible although, in one case from what I can tell, it allows for observations to be written in a simple way -- adding more structured detail simply makes the data more useful in terms of parsing and searching.
My own little experiment is about making it as simple as possible. I'm starting to re-write the electronic copies of my logs in XML and have written an XSL file to transform the XML into HTML (in this case transform it back into the HTML form that I initially wrote when I started keeping copies of the logs). You can see an example XML coded log here
(that URL is going to be good for around the next 24hrs). This is the XML version of this
and the original "source" for that log is here
. As you'll see from all of them, the XML version is better structured than the HTML written one and I can get the HTML version back out of the XML (using this
(if that URL gives a problem with your browser you might want to do a download from the link instead and then look at the file -- again, the URL's good for at least 24hrs) XSL file in conjunction with Sablotron
Moreover, once I've got all my logs in XML I can do all sorts of other things with them. Already I've got a more automated system for building the general and Messier indexes (using the index lists that are written into the XML files) and adding searching and the like would be easy enough in the future.
jeff.stevens wrote:It will be very interesting to see where you take this project though, keep us posted.
Will do. As you can see from the above, I've got a format that works for me right now, is more useful and which allows me to carry on using the framework I've already got available on my website. The nice thing about using the XML approach is that if/when I re-write my website (and I aim to do that in the next year or so, I've had this one since around 2000) I don't have to go back and re-work all the logs to fit the new format, all I need to do is run all the XML files via a different XSL file and the job is done.
I should stress that the point of my own little experiment here isn't
to design and propose a format that could be useful for anyone (although anyone wanting to make use of it is more than welcome to do so) and which could form the basis of a shared log site (although, at a push, I guess it could). The point of my original post was to see if anyone else had played with this problem and to see what their solutions looked like. I'm still considering the idea of using one of the existing projects I mentioned in my first post and I'm also hoping for some useful feedback from someone like the BAA (if any feedback from the SPA was available I was hoping that it might be available via this specific forum). The nice thing about what I'm currently playing with is that should a better format become available it should be a lot easier to convert my XML files over to it because I'll already have the electronic copy of my logs in an easy-to-read(-for-the-computer) format.