I started seriously following web trends a few months before 2008. Now 2+ years later, I’ve deleted 500 bookmarks from my Delicious, almost a couple hundred+ tagged items from my main Reader account, and another couple hundred labeled as TODO from my browser bookmarks, stored by XMarks. Now, hours later, I still have another couple hundred items in my ReadItLater list, although somewhat sanely manageable and accessible. And I’m fanning the flame by writing a long review, hoping some upstart doing websites out there will find use in it.

Overall, I feel like all web news, lessons, and exhibitions was useful. I learned lots stuff good. But I spent much time and still missed some golden opportunities. On several occasions mostly in Delicious, I had duplicated entries perhaps only months apart. A number of the items I had as TODO were horrible write-ups on WordPress and jQuery “tricks” that were (in terms of screen estate) 20% awful UI, 40% yes-men comments, 30% ads, and 10% content. Only now do I realize how shallow those authors’ knowledge were, as I eventually outgrew the usefulness of my gathered resources. So yes, I should have organized better (more on that), but there is a definitely content quality issue with sites out there. Oh, and I’m also pissed FFFFound has no RSS feed.

Seems like posters of late don’t take the time to write and grow content but instead just put up link bait to worsen the quality of RSS feeds. Delicious Popular, back in 2007, did not look as it does today. Back then, since the web was less spammed up by WordPress strip malls, the content gathered by (what I’m assuming is and has been) an automatically ranked feed of popular pages on Delicious based on times bookmarked. As a result, the content was quite good, and I learned a lot about mashups, and startups, and CSS standards, and even design patterns and separation of concerns. Back then, there were no Noupe, Cats Who Code, etc., and point is, now my attention has shifted to Hacker News and a selected group of consistently good bloggers writers. Still, little by little the WordPress spammers are creeping in. I really don’t know how they do it, and I don’t think HN is very automated…

For the past few months, I've been fighting this spam with better organization. I got the PostRank userscript for Reader that can at least visually show me which posts are truly more popular, based on a fairly good algorithm. I decided to stop using web interfaces for browsing bookmarks. One, because the tags feature in Reader is limited and difficult to browse. Two, Delicious’ website, browser sidebar, and assumptive other services are abysmal with managing a huge collection of bookmarks. Their bookmark editing features are worse: granular editing cannot be done in bulk, as their bulk editing feature lacks key functionality (that require little effort beyond some jQuery, or in their case, YUI).

Browser bookmark management, especially in Firefox, is way more advanced. It can quickly show hierarchy through nesting folders (vs. the Delicious tag+tag approach). It has separators—translated as meta categories—for dealing with over-nesting folders. It is convenient. Use case: click drag the favicon to a folder in the bookmark bar, then separate as needed in the menu; access is a click away, and editing and management are almost better than Finder/Explorer. It includes favicon support and compact lists. Plus, XMarks syncs bookmarks across browsers using SSL encryption (yes, most bookmarks (at least mine) need not be public). Don’t get me wrong, Delicious is a fine service and website. It just isn’t nearly the best bookmark handler.

Admittedly, when I first started on Delicious I added every semi-interesting or useful page to my account, each attached to a dozen or half-dozen tags, and ended up with a huge, flat, and unstructured pile of links. I realize today a lot of the software sites I don’t need to bookmark; I’ll download it and try it out, and there’s no huge time cost there. Interesting, but long, articles are a big time cost and break the feed-pruning flow, so I only decide after skimming whether or not to queue it to ReadItLater, which comes with various integration points: bookmarklets, Firefox plugin, RSS (which ties into Reader and its feed bookmarklet). The only things I save now are seminal articles, tutorials, and documentation to keep; and libraries, documentation, and work I want to dissect in detail, and web tools whose urls are hard to remember. Basically, anything that takes an hour or more to sink in, or that I want to revisit again and again. Meanwhile TabGroupsManager acts like a temporary work area for storing second-class bookmarks and can export them nicely as bookmark folders.

There’s two points I want to make as I end this reflection. One is centralization is underrated. People want to share here and post there and jump from Youtube to Facebook to browser to iPhone. But we’ll use more and more apps, and we’ll realize how important centralization, via standardization and organization, really this. Dropbox proves this as a bridge between multiple PCs and a Webby nominee for this year, so does Digsby, with its multiple social, chat, and email services, and its CNET 100 award . They are services acting as bridges and gap fillers. They save me time and at the same time allow me to do and learn more. And so, my solution for bookmarks chaos is to rely on the browser as the core bookmark manager, and to use various web services to do the syncing. Of course when the need arises, I will share things I actually feel are useful for first my immediate Delicious network and then the rest of the world. It’s important to realize the prospective bookmark's level of importance in context of your network and workflow. And that’s the second point.

Why did I do all this for a pile of “trivial” bookmarks, able to fit in a small and dispensable file? Simply because I learned most of what I know from other people’s bookmarks, and I want the web to improve in organization and quality. In my case (which I’m sure is not unique), it actually got to the point where I got so confused with the quantity of links I couldn’t find the crucial ones I needed. It was as if I had no bookmarks, as I just ended up googling everything and relying on my tab sessions. Because ever since middle of 2008, thousands of WordPress articles starting showing up in my feed. Even after skipping 90, each of the remaining ten still looked so useful in the 5 seconds I had to skim it before filing it away forever, only to uncover it again months later and realize the documentation I poured myself over while working already taught me many times more.

Readers beware, lots of crap out there.

Update Half a year since writing this, and much of the technical details above has changed, but my overall setup still follows the same structure. I no longer can recommend XMarks because they've pretty much closed shop. And although they might receive some funding or improve their business model, in their blog post they nailed something I failed to see: native is better. Google made native sync, and I primarily use Chrome now. That's right, no more jumping back and forth from Firefox to Chrome, at least for a while. Although not having bookmark separators is a bit of a dealbreaker for me, in Chrome I can actively manage memory and have a generally stable environment (with the exception of web inspector in certain instances freaking out over javascript errors and crashing the tabbed process). 

Also, with an iPad I now have access to other readers like Flipboard, which as of writing has Google Reader support. As a result, image-based feeds and newspaper feeds are much more convenient and fun (especially to go through why lazying around under a blanket).