I’ve been running the annual Top vBlog voting for the last 6 years and as the process is very time-consuming I’m always looking for ways to improve and streamline things. I was recently reading a post from Rene which had some ideas including using stats to determine who the top bloggers. I don’t really like that idea as it then becomes more like a video game to see who can get the highest score and people will do anything to get it.
There is no doubt that Top vBlog is a popularity contest but isn’t pretty much everything that we vote on in life such as the Presidency, Pro Sports All-star Teams, America’s Got Talent, etc, that’s just the nature of what voting is and that’s really OK as long as it doesn’t get out of control. If we start trying to rank things purely based on statistics you lose the most important element, what the people think and want. The fact of the matter is that the stats will mostly line up with the voting results.
When I open the voting I try to give two key pieces of guidance, first make sure bloggers understand that while this is a popularity contest the intended voters are those that work with virtualization technologies and not everyone in your office or who you are Facebook friends with. Second for voters to be fair and to take the following into account when choosing their favorite blogs: Longevity, frequency, quality and length, you can read more about those areas here.
So I probably won’t be radically changing things but I am open to suggestions to try and make things better. One suggestion that I did like from Rene was on the post counts for the year of each blog. Right now I display blogs in the voting survey alphabetically to keep them all on level ground, but I bold the previous years top 50 so they stand out. I like the idea of sorting them by number of posts so the most active bloggers are near the top and those that hardly post at all are near the bottom. Because there are so many blogs now this will help get the most active bloggers better visibility on the survey. I’ve looked around to see if that could be automated by scrapping blogs and parsing the archives to get the number of blog posts but haven’t found anything that will do that. Andreas manually did this by going to each blog so I can probably leverage some volunteers to divide and conquer and do the math for me.
As far as using other metrics like like if they wrote a book, presented at VMworld and how many people follow them on Twitter I don’t find those as being a good measuring stick to how good and popular a blogger is and it penalizes a lot of the bloggers. So if you have any suggestions put them in the comments, one crazy idea I had was to make it into NCAA Basketball play-off style with the bloggers all facing off to the make it to the Sweet 16 and then the Final Four, but while that sounds cool I don’t think that would play out well. Here are some other things I’d be interested in having feedback on:
- Should it be open it up to selecting more than 10 blogs?
- Should categories be optional?
- Are there any additional categories that might be interesting?
- Should blogs be displayed on the survey in # of posts order from most to least?
- Any other ideas or suggestions you have!
I am also looking to implement a database-backed system instead of the current WordPress tables model which is a bear to update, this would make it more of a directory style and would be much easier to work with. I’m looking into a few options for that right now.
Sound off in the comments and look forward to hearing from you.
>Should it be open it up to selecting more than 10 blogs?
Yes… but it become more difficult handle it. IMHO more than one choiche for categories could help.
>Should categories be optional?
No… Now if you don’t want you can already say other.
IMHO more than one choice could be useful.
>Are there any additional categories that might be interesting?
Storage is there… Network virtualization should be there.
Rene suggest also other virtualization platform… but become more complicated (IMHO other content are needed for them).
>Should blogs be displayed on the survey in # of posts order from most to least?
IMHO they should be displayes random (you already have the top in bold version).
Should be nice have a form or widget to start the vote also from a blog web site (but could be complicated handle how manage this)… something to “put in basket” in order to allow people to visit the blog and select which they like.
Thanks Eric for the phenomenal amount of work you do on this.
My suggestion would be to not just have to rank blogs on a top 10 basis. This I believes focuses the attention too much on only the top 10 and makes blog number 59 for example which may never have a true chance at top 10 stardom be recognised as a useful blog. Perhaps a secondary list of 10 blogs you also found useful, maybe not even ranked by number. This would allow voters to give their true top 10 for the overall industry but perhaps also give focus to a bunch of say training or vRealize blogs that are important to them but would never displace any of the top 10. The votes for this category could influce the rest of the rankings is a more meaningful way than just the top 10.
There’s no reason to complicate it too much as well but also some of Rene’s automation steps could ease the burden on signing up blogs and crunching the results.
Eric, kudos for the great work you do with the vCommunity. The Top vBlog List is a great resource used by everyone.
My two-cents to your questions:
* Should it be opened it up to selecting more than 10 blogs? Possibly, but the voting engine should allow a minimum of 1 blog selection.
* Should categories be optional? Yes, let people vote on their categories of interest.
* Are there any additional categories that might be interesting? Architecture Design & Strategy, Converged/Hyper-Converged Infrastructure, DevOps.
* Should blogs be displayed on the survey in # of posts order from most to least? Yes.
* I hope a Developer(s) can volunteer time to assist you with the automation of your Top vBlog process.
* If you need assistance with manual data collection and entry (eg. # posts), count me in.
I would like suggest that some certain criteria be added to eligibility for being considered for the top X blogs.
First and foremost – continuity. i.e there should be a minimum number of blog posts that have to have been posted in the the year (and not all in the last week/month/quarter) to be considered eligible for participation – I think that a 1 post a month is more than fair.
I do not think that publishing 6 posts in the year 2014 quantifies and there are at least 2 bloggers I know of in the top 50.
Regarding the questions
* Should it be opened it up to selecting more than 10 blogs? Yes
* Should categories be optional? A definite yes
* Are there any additional categories that might be interesting? Cloud, Networking, Automation (this should be the one to replace scripting) * Should blogs be displayed on the survey in # of posts order from most to least? Perhaps group them and then sort alphabetically – i.e. 1 per month, 2 per month, three per month….
Please feel free to reach out if you need any help
Yeah I tend to agree with the minimum, I was thinking of putting it at 10. Right now I exclude any blogs that do no post in a year. I think doing that would knock off at least a few dozen blogs.
I was also thinking of changing it to pick 12 as with so many good blogs its hard to pick only 10.
The last weeks I have updated my “vBlogs 2014 by the numbers” chart with new fields as some people asked for them (eg. country, team or personal blog,…). I will publish it in the next weeks at running-system.com.
But I have the same problems like you – wordpress tables are tricky to work with – a database would be better. So I am working at a database model for this at the moment (not so easy, because time is rare)
Maybe this database would make your life easier, too?
Manual collection of post numbers: I did it this january and with some tricks it is relatively easy. So if you need help here – drop me a mail or DM at Twitter. I will do it definitely for running-system.com again next year – so these numbers will be available.
Thanks Andreas, sounds like you’re volunteering to do it next year so that will save me some work 😉 I think would I decided on was to not rank the votes by number of posts on the ballot but instead exclude any blog with fewer than 8 posts in 2015.