2013: DrupalCon Portland Session Notes & Thoughts

DrupalCon 2013: Summing it all up…

It’s always hard to judge DrupalCon’s effectiveness, mainly because I am always confused as to how im measuring it.

  • Did I enjoy meeting up with new and old colleagues? Absolutely.
  • Were the sessions amazing? Not enough of them were.
  • Were the BOFs amazing? They would have been better if the more popular ones were in bigger rooms (EDU, Aegir)
  • Did I enjoy DrupalCon itself? Ask again later.

 

The Good:

  • EDU Unconsortium
    • I’m hopefully going to be able to help with the new attempt to provide a central place for all Educational Drupal users – http://edudu.org
    • It’s an idea started by colleagues at Stanford, UC Davis, and UC Berkeley.. and now a few of us are going to lend some time to help get it off the ground (and to continue making it a useful resource)
  • EDU BOF
    • I’m putting this in the “Good” category, because it’s always good to hear what challenges or successes are happening at other institutions, no matter the size.
    • The take away from all educational-centric sessions, is that NC State remains in the dark ages in terms of brand consistency. It actually emitted open laughing from the audience in 3 different sessions when the premise of a University not attempting to implement a single, unified web brand/presence was mentioned.
    • It took several conversations and explanations about how NC State is “decentralized” for colleagues to understand why there is seemingly no push from the top to standardize on a platform, or at least a single look and feel.
  • Aegir BOF
    • @theanarcat– one of the creators/core maintainers of the Aegir project attended, and graciously spent time answering questions and discussing things with us.
    • Aegir should be the gold standard if any one is running a Hosted Drupal Service, and yet some people dont even know about it (see “The Bad” section below for more on this).
  • New tools that i didn’t know about
  • Helping colleagues
    • For the first time, i was able to share a lot with my colleagues (both old friends and new) regarding the Hosted Drupal Service at NC State…. as well as more compartmental help regarding individual components (Aegir configuration, Views tips, Accessibility tips to name a few).
    • I am glad im at the stage of being able to give back, rather than just being a taker of knowledge from others.

The Okay

  • BOFs
    • The amount of BOFs on offer were great, but they were all set up through the Unconference “White Board” method… meaning that the board filled up half way through the 2nd day… and who knows what other topics didn’t see the light of day.
  • The Keynote
    • Dries started out by saying we “need to do good” with our time, and our usage of Drupal. IE: to help people, to advance opportunity, to embrace the open-source way of doing things. And i totally agreed, and was on board.
    • But, then he proceeded to inextricably link his perception of “success” of Drupal with the number of websites using it. I think that a better product will drive more adoption of the product, whereas the drive from Acquia (the business-arm of Drupal, run by Dries) seems to be “lets just get everyone in the world to use Drupal and then everything will be peachy.
    • If they fix the editor experience (which is on the cards for Drupal 8), then i think there’s a chance Drupal may become a more attractive option for those people (including me) who use WordPress for more simple blog-ish sites. Hell, this post is on a wordpress blog… because it takes less to maintain than a Drupal site if all im doing is blogging.

The Bad

  • The conference website
    • Why didn’t they just use COD (http://drupal.org/project/cod)?
    • What the crap was the scrolling table thing for the schedule, that was almost unusable on a mobile device?
    • Why should i have to use an app made by a 3rd party just to navigate to which sessions i want to attend.
  • The black-listing of Aegir-related sessions
    • As i said above, Aegir should be the gold standard for managing large drupal environments….. but not everyone knows about it.
    • Aegir sessions were NOTICEABLY absent from the schedule. None were approved. Zero. Zilch. Nada.
    • All we got was a BOF that almost didn’t make it on the board. And whereas the BOF was great, it should have been more. It should have been a session in its own right, maybe even a panel, in a huge room that im sure would have been packed.
    • Why is it seemingly blacklisted? Because Aegir represents a way for people to manage large scale services themselves, and therefore would not require the help from the sponsors of DrupalCon. And that’s what is disappointing: the slow creep towards Drupal being a wholly owned and operated system for businesses to make money from, rather than a true open source collaborative community. Acquia has the right to make money from Drupal just like any of us do… but the lines are becoming blurred between Drupal the open source software, and Drupal the business (aka Acquia/New Media/Pantheon/Commerce Guys/the list goes on)

Wrap it up

  • It was worth attending. That’s for sure. But more now than 4 years ago…. the sessions at DrupalCon are becoming less important for me.
  • Sure, the people are the best thing about attending a conference, but you at least expect sessions that drive conversation, or enquiry, or collaboration. DrupalCon Portland didn’t do that for me, but hopefully im in the minority.
  • The next DrupalCon is in Austin. In June. Wow, that’s going to be hot. Not hawt, like awesome dude!…. but hot as in melt your face off hot. And of course im looking forward to it.

 

Very rudimentary notes that i took in some of the sessions are in this Google Doc:

Go Links: More statistics

For the fans of statistics, this post is for you. All statistics compiled as of 9:30am, May 13th 2013

General stats

  • Number of unique links: 10,283
  • Number of links with zero hits: 3,890
  • Number of links with at least 1 hit: 6,393
  • Number of links with at least 100 hits: 1,098
  • Number of links with at least 500 hits: 336
  • Number of links with at least 1000 hits: 195
  • Number of links with at least 10,000 hits: 27

Top 10 Links by total hit count

  1. go.ncsu.edu/kiosk
  2. go.ncsu.edu/class_search
  3. go.ncsu.edu/print
  4. [private dept link}
  5. go.ncsu.edu/curricula_requirements
  6. [private employee link]
  7. go.ncsu.edu/myschedule
  8. go.ncsu.edu/ncres
  9. go.ncsu.edu/advance
  10. go.ncsu.edu/mygrades

Hit Count by Year

  • 2010: 16,548
  • 2011: 342,256
  • 2012: 1,023,580
  • 2013: 1,295,782 (as at May 13th)

Hit Count By Month (regardless of year)

  • Jan: 296,894
  • Feb: 269,962
  • Mar: 352,480
  • Apr: 461,235
  • May: 325,707
  • Jun: 65,866
  • Jul: 89,246
  • Aug: 167,698
  • Sep: 128,535
  • Oct: 177,394
  • Nov: 170,601
  • Dec: 172,773

Hit Count by Day of Week (regardless of year or month)

  • Sunday: 474,615
  • Monday: 471,362
  • Tuesday: 481,418
  • Wednesday: 464,474
  • Thursday: 381,784
  • Friday: 190,266
  • Saturday: 214,796

Owner stats

  • Number of users with at least 1 link: 734
  • Number of users with 10 or more links: 153
  • Number of users with 50 or more links: 18
  • Number of users with 100 or more links: 8

Go Links keeps on ticking… Surpassing 2.5 Million Hits

Go Links launched in October of 2010, and between then and September 27th 2012 it recorded slightly over a million hits. That’s a million hits in 728 days (give or take a few). That works out to be approximately 1300 hits per day for the 7503 links that existed.

Today, on May 13th 2013 (229 days later), Go Links has surpassed 2.5 Million Hits. At the time I started writing this post, the hit count was 2,673,926. That’s an average of over 7000 hits per day, every day since September 27th 2012…. or almost 2800 hits per day since Go Links was originally launched in 2010. Now there are 10,281 links (2778 added since than 9/27/12)

You can drill down into the statistics here if you’re interested.

What does this mean?

For me, the Service Owner and Developer, it means that I need to do some analysis and upgrades so Go Links doesn’t crumble under the weight of its popularity. Over the next couple of months, I will be rewriting pieces (large pieces) of the code to be much more efficient….which is something any 3 year old application needs to undergo anyway.

It also means that some underused or ignored functionality will be removed. These decisions will be based on data and actual usage, rather than any subjective opinion. With this many links, and 733 unique link owners, it’s important that we make supportable decisions.

Any changes already decided upon?

One change that is almost inevitable, is to remove the “Theme” feature, so that all links are more simply directed to the target URL. The theme feature, while necessary in the beginning to help people understand what Go Links were, and to provide feedback mechanisms, cause a couple of problems: (1) increased load on the server, (2) increased complexity inside the application code, and (3) Mobile & Printer formatting problems.

When will any changes happen?

I am actively working on some of the changes behind the scenes, but there will be no visible changes to the production Go Links service without a larger announcement on sysnews. That announcement will be cleared with the Change Advisory Board (CAB) and a date will be set (hopefully sometime this summer). Go Link owners will not have to worry too much about down time, as the changes will most likely happen overnight, or at least during non-business hours.

Questions/Comments/Concerns?

If you have any initial questions about Go Links, or anything in the post above, please contact me at oitdesign@ncsu.edu.

And now, at the end of this post, the hit count is 2,677,093 (3,167 hits in about 30 minutes…. i guess the start of the day is the busy time :-)

 

DrupalCon 2013: Portland … pre-thoughts

In just over a week, i’ll be attending my 4th DrupalCon in 5 years… this time in Portland.

Things I’m looking forward to:

  • Education sessions, specifically BOFs
  • Aegir conversations
  • Migration paths for Drupal 6->7, 6->8, and 7->8 (because we’re still on 6, and i dont have a migration plan ironed out yet)
  • Catching up with friends and colleagues from North Carolina, specifically my friends at Appalachian State, who gave me the premise and more than a little help for the Hosted Drupal Service.
  • Observing hipsters in the natural, original habitat.

I’m sure there will be more than that to pique my interest, but that’s my initial plan. I am unsure if anyone else from NC State is going this time, but if anyone wants me to seek out particular information while i’m there, let me know.

I’m planning on blogging/writing up some session summaries, or maybe just a day by day recap too, just like previous years.

I'm going to DrupalCon Portland

Making links more accessible (new Drupal module available)

Recently, our awesome University IT Accessibility Coordinator launched a tool that allows NC State website owners to scan their sites for accessibility issues (and usability, broken link, W3C Standards issues too). And after scanning many of the sites that i own or developed, a trend emerged: The top accessibility error (in terms of number of instances/offenses) on all my websites was:

     “Avoid specifying a new window as the target of a link with target=”_blank”

Now, while the debate will rage for many more months or years about whether it is right or wrong to forcibly open a link in a new tab or window, the accessibility issue is unquestionable.

Huh? What’s the problem?

The premise is that you need to provide information to the user before they click on a link, that the link will open in a new window or tab. Otherwise, when they click the link and end up on a new window, their back/forward buttons won’t work as expected.

The Non-Drupal Solution:

If you’re running a regular non-cms website, then you’ll have to fix all your links manually… or you could add a little jquery+css to do it for you: http://jsbin.com/inuyop/2

Look at the source code for that page, and you’ll see that it searches the page for any link that opens in a new window, and adds a class to it, then adds a <span> with some text, which alerts the user on :focus, :hover and :active that a link is going to open in a new window.

Implementing this will solve your accessibility issues, but your fix won’t be reflected in the stats on the Accessibility Scan for NC State. The scanner doesn’t pick up the javascript, and so you won’t get credit in your ranking.

The Drupal Solution:

In Drupal, you can take advantage of “Filters” which run before the page is rendered, thereby changing the actual source code that is displayed (and scanned by the Accessibility Scanner). I developed a new Filter (not available on Drupal.org yet) that packages up all the functionality into a nice little module that you can then add to whatever input format you want. This means you can enable/disable it on different input formats, for different content types or whatever you want.

Here’s the module for download:

http://drupal.ncsu.edu/features/project/accessible-target-blank-filter

… or you can contribute on Github:

https://github.com/usaussie/drupal6-module-accessibletargetblankfilter

Right now there is only the Drupal 6 Version available, but i am planning to provide the Drupal 7 version shortly. As an added bonus, this module will add a title attribute to any matched links that don’t already have a title attribute.

This module is enabled on http://oit.ncsu.edu (hover over the “follow us on twitter” link on the right sidebar) as well as all the Hosted Drupal Websites at NC State.

Credit goes to Greg Kraus for his explanation and assistance, and you can check out his information about this accessibility issue here:

http://accessibility.oit.ncsu.edu/training/accessibility-handbook/link-new-window.html