Moving to a Self-Hosted Bookmarking Service
There's a hint that bookmarks are going away. Services like Delicious that used
to provide great web-based book marking services have closed down. In their
place are the likes of Instapaper and Pinboard and the like that break stuff
into different types. Instapaper to store sites that you want to read later,
Pinterest for images you like, and so forth. That doesn't work well for me,
though. I want everything in one place. That's also why I decided to try a
couple of web-based products. I realized that I had bookmarks scattered over
three different browsers, on three different computers. How confusing can it
get?
Why Self Hosted?
With a variety of services already available on line, why do I want to
self-host? After all, that's a lot of work to get things set up and to keep
things running correctly. It would be much simpler to use a series that
already exists.
The primary reason is safety. That is, These services like to shut down without
a lot of notice. I don't want to collect lots of bookmarks only to have the
service shutdown, taking my links with them. I want them to persist. By hosting
my own, the only risk is that my computer will die. I back things up for that
reason.
Another reason is privacy. It's not so much what I'm bookmarking. I don't
really care if you know what I bookmark. But I do care that you don't try to
market me, or sell my information to someone else who will market me. When using
another service, that risk is real -- they do have to make money. But, by
hosting my own, I know that can't happen. And, if it does, I get the financial
proceeds :).
What Services Am I Using?
As I said, these services are falling to the wayside, so there aren't a lot of
scripts that are still available and being actively maintained. I found 3:
SemanticScuttle, Unmark, and Shaarli. I installed all three so that I could try
them out to see what works for me. I'm not sure which one I'll settle one. But,
I'm already sure which one I won't be using.
Before I get into any specifics, though, let me give you a little background on
what I prefer. And, where I'm likely going.
Tags have been around for a while now and I've rarely used them. I'll touch
on that in another post. I'm a bit old-fashioned and a little linear in my
thinking. I like being able to save a bookmark to a specific folder so that I
can quickly find the list of bookmarks related to a specific subject. Yes, I
know tags allow that, and more and, again, I'll talk about that in another post.
but bear with me for right now. I wanted something that would allow me to
import my current bookmarks, maintaining the current structure. None of the
available scripts allowed this (there is one, but it tried to be too many
things, I've tried it in the past, and its interface is hopelessly outdated). I
also wanted
to be able to quickly and easily remove duplicates. And, of course, it had to
be easy to access and use. Let's take a quick look at the three
I installed.
Unmark
As I mentioned above, none of the three I found supported folders; all supported
tagging, so I ended up going through and tagging all my bookmarks. Unmark held
a lot of promise. I read a lot of good things online about it. IN practice, I
wasn't impressed. The biggest issue is that Unmark wouldn't import the tags
associated with the links. That meant I'd need to go through and re-tag all
those links. That didn't impress me. Moving forward, of course, it's less an
issue. But, initially, at least, Unmark failed. I've also has some issues with
Unmark not loading correctly after signing in.
SemanticScuttle
SS imported bookmarks and tags without issue. So far, too, it's been easy to
use. I'm still using this one to try to see what it's limits are. So far,
though, I'm impressed. The one drawback is that it depends on a MySQL database.
Now, I've worked with DBs for years and don't' automatically discount a script
that uses one. But, I think for my uses, a DB is simply overkill. It requires
a lot of resource for little added benefit.
Shaarli
Shaarli and SS are running neck and neck in the which one does Kevin use race.
From the user perspective, the two are identical. And, like SS, Shaarli
imported tags without issue. Sweet. The other thing about Shaarli that I like
is that it doesn't depend on a database. It uses a simple XML file to store its
data. That not only makes backing up the data dead simple, it also means
minimal resources. I could literally copy the files to a new computer and just
keep going. That is an awesome feature. Shaarli also allows you tom make short
notes, or even use it as a pastebin kind of service.
Which One Do I Recommend?
I don't know yet. Unlike some folks that spend a day or two playing with a
service and then making a pronouncement about how good (or bad) it is, I want to
work with them a while to see which fits into my work flow best.