Travel Blogging Is Hard

Background

Ever since I broke off blogging about my time in Rome once things got too busy, I’ve had the expectation of catching back up and finishing what I started. I’ve even promised people that I’d finish, and been saying that I would once I got back stateside.

However, I had an online class to wrap up, some other travel to deal with, and responsibilities related to job-hunting (it has begun!). And the lazy troll inside of me made me take some time off too. But if I’m being honest, the real reason why I haven’t just gotten back into it (even though closer proximity to the trip would help me make better sense of my often-abbreviated notes) is simply that it’s a lot of work! Blogging for me before this trip has been demonstrably haphazard, and while I’ve often had aspirations of making it a more regular thing, it just hasn’t happened. But it was also always a more as-fancy-took-me sort of thing. And typing up random musings, it turns out, is a whole lot less work that working on organized and informative posts with good pictures. What do I mean?

Steps in my travel blog process

Typing up raw class notes

I took notes on my Microsoft Surface tablet for the duration of the trip, making use of the Surface Pen and a well-designed tablet grip (highly recommend). I prioritized my notetaking, and probably filled the equivalent of 5 or 6 pages every day. Since I typically had to write fast, my handwriting suffered, and is hard for even me to decipher in places.

What’s more, I used a fair bit of abbreviation in my notes. Nothing too systematized or formal (like Gregg shorthand, for example), but just normal abbreviations like “w/o” for “without”, as well as context-dependent abbreviations. Some of these that come to mind are “A” for “Augustus”, “FoT” for “Forum of Trajan”, and “Pp” for “Pompeii”. There would have been absolutely no way for me to take down the volume of information I did if I did not use abbreviations heavily. However, they do make typing up the notes – “post-processing,” if you will – a bit more effort-intensive.

Matching like content

Very often, my professor’s lectures would bring up a topic, and then circle back to it later. Rather that keeping all of the information on something together in my notes (which would require copying and pasting tablet handwriting around on the fly and leaving space to fill in more notes later), it was far more practical for me to just take things down as they came, and do my best to draw arrows as appropriate. To get information grouped together, I thus have to make an initial pass-through of my notes, following the arrows to rearrange information such that like information is mostly grouped together.

Even after doing an initial pass like this, I typically have to spend some additional time doing this on a wider scale. It turns out that it is not uncommon for me to have notes about the same general thing come up even two or three vertical screens apart – far enough apart that there are no arrows in the notes themselves. Thus, to actually get like content grouped together, I have to read through the notes once or twice and make sure I have things arranged properly.

Adding structure and organization

Even after the notes are grouped such that like content is together, I still have to add section names (headers for the Table of Contents to automatically pull in – the raw notes contain nothing of the sort, since they are composed of just information without organization), and also arrange the sections in a hierarchy that makes sense. I not only need to decide which hierarchical relationships make sense, but also what order the hierarchy ought to have in general: what information should come first? To be able to do all this logically, I have to sit there and make sure I actually understand the information in the notes, and understand how it all relates. Doing this takes more time than one might think initially.

Selecting the best pictures for particular shots

I took a lot of pictures on this trip. Probably a couple thousand all said and done. For each shot I actually wanted, I took 4 or 5 pictures. Part of this is because I had nowhere near the time to set up the aperture, ISO, and shutter speed to any meaningful degree (I was lucky to get 10 seconds to photograph something), and while the fully automatic selection of these things did quite well most of the time, the values were dependent upon the focal area, and if you got unlucky and happened to have the frame directed in such a way that the central focus was pointed at an area of very high or low light, the picture would turn out underexposed or overexposed on the whole. It was faster and more efficient for me to take a rapid set of pictures of something than try to set up a single shot that would have a better chance of being guaranteed to turn out. Statistics takes care of the rest.

While this paradigm served me well (I can only think of a few instances where I haven’t been satisfied with any one of a group of pictures for a particular shot), it has cost me a lot of time in post, since I have to manually go through 5-ish pictures for each and every shot and pick the best of the lot. Sometimes this is easy enough because all the pictures turn out and I can mostly arbitrarily pick one, but sometimes I really do need to sit there and compare all the pictures. This is mostly the case in shots that had iffier lighting to begin with.

It’s not just lighting either. The point-and-shoot I was using didn’t have a level function (or if it did I missed it when looking for it), and so if I wanted to try and keep an edge straight in a shot (which is very common), I had to do my best to keep a level hand and take enough pictures for the shot that one of them turned out with respect to both edge orientation and lighting. This too requires manual comparisons to sort out. There were some ceiling frescos that I took upwards of 30 pictures of to try and make sure I would get a picture that was aligned, with appropriate lighting, in good focus.

The combination of all these things means that just to get a base set of possible pictures to use in any given blog post, I have to do a lot of initial legwork. Since I had my camera shooting dual RAW and JPG for all the shots, I do all of this initial deletion of unwanted shots with things unseparated: I delete the RAW and JPG versions of each unwanted shot at the same time.

Organizing and naming the pictures

Once I have the best pictures chosen for each shot, I split the RAW and JPG files into separate folders within a parent folder for the given day, and name all the shots so that I know what they are. Some things are obvious, but other things require me to refer to my notes to figure out what the pictures are of. This process takes some time, but it will ensure that I can continue to enjoy my experiences on this trip even years from now when I would not otherwise be able to tell one picture from another.

Working on the pictures

If I decide that I need to do any work on the pictures I’ve selected, I’ll open the RAW files in a photo program and do some basic corrections. I’ve mostly only done color rebalancing for shots where incandescent lighting is giving things a reddish-orange tint. Fixing white balance is usually just a couple clicks, and then it’s simply a matter of saving the RAW file and overwriting the initial JPG with a corrected version derived from the fixed RAW file.

There are some other shots I’ve done stuff too (stained glass comes to mind – I mostly just gave up and accepted that I’m not a good enough photographer and/or didn’t have enough time per shot to get good stained glass pictures on this trip), but I’ve been pretty satisfied with the stock photos coming out of the camera.

Selecting the best pictures to actually use

Not all the pictures I took end up in the blog posts because some shots are just better or more relevant than others. It’s hard to put a definite number on it since the number of total shots and shots used varies from day to day, but I’d estimate that I typically use under two-thirds of all the shots I took for a specific day in the blog post for that day.

As a general rule, I don’t include pictures of things that I don’t have any or many notes on, just because having a picture without much context seems much less interesting to me on the whole. This is probably the primary criteria I use to narrow down which pictures I use, but I also tend to just choose the pictures that I think are best from a photography perspective. While I’m far from an experienced photographer (AKA I only know vaguely how to distinguish between good and bad photographs), I really don’t see a reason to include shots that are of lesser quality from my perspective. Maybe they are boring, or simply don’t have as much experimentation with perspective, angles, or foreground and background as some of my other shots. I also try to avoid including a bunch of shots that are pretty similar to one another. I’ll be the first to admit that there is a lot of subjectivity in all this, but hey, it is my blog after all.

In line with the importance I put on having notes related to the shots I use, I typically decide where I will put shots in relation to the text at the same time I am choosing which shots to use.

Actually inserting the pictures

I use Emacs’ Org mode to write all the content on my website. To automate the insertion of pictures, I use a regular expression to turn a list of file names into a list of appropriately-formatted picture links (with the pictures themselves linking to their raw forms directly to make them easy for people to download – remember that stuff on the blog, including pictures, is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License), and then paste these picture links into the appropriate places in my writing. If you’re interested, here’s what a regular expression might look like for a file of enter-delimited filenames to turn into picture links (the name of the blog post changes in the regular expression depending upon what post I am working with):

RE Search: (.*)
RE Replace: [[https://www.steventammen.com/posts/06-10-19/$1.JPG/][file:/posts/06-10-19/$1.JPG]]

After inserting the picture links, I copy the JPGs from the picture archive on my computer to the directory under static/ for the given blog post, and then spin up my localhost webserver for my site to proofread.

Proofreading and publishing

Once everything is pulled together, I’ll give the post a couple reads to catch typos and make sure everything flows well and looks good. After making any appropriate changes, I’ll publish my draft as a post (giving it appropriate metadata). Finally, it’s a git add, commit, and push, and then the post goes up in the GitHub repo and Netlify handles the rest.

Takeaways

Well, there you have it. Travel blogging, at least how I’m doing it, is an effort-intensive process, and this explains in part why I’ve been slow to get back into it on top of my other commitments.

At some point I’ll probably also add corresponding readings in book sources for each post (since our professor gave us a nice list of readings for topics corresponding to the schedule), and also add material to various posts as I get around to incorporating the work I did on my 60-page or so study guide for our final exams. I don’t just want to post the study guide due to ethical considerations for future semesters, but I did work hard on it, and a lot of the information I synthesized in the study guide goes above and beyond the information of my notes. These things will happen eventually I suppose – more steps in the process!

 


comments powered by Disqus