When volunteer speakers submit a session for the PASS Summit conference, volunteers on the Program Committee have the tough job of ranking them and picking winners. In the past, this has been done completely in private, and speakers had no idea why their sessions weren’t accepted. This meant they couldn’t work to get better.
Thanks to requests from the community, that’s changing this year! Amy Lewis, the Board of Director member responsible for the process, writes in her blog post announcement:
“If you submitted a session proposal for Summit 2014 and would like the comments from the abstract review team, please email program@sqlpass.org and we will send you the comments that are available for your abstract. Please note that, as stated above, not all abstracts have comments, and the comments are related to the abstract only and not why the session was or wasn’t accepted.”
This is awesome news, and it’s a great step in the right direction. The second step would be telling speakers why the session wasn’t accepted, which can be for a totally different reason. (You’ll see why that’s important in a minute.)
Developers: Who Needs a DBA? (Accepted)
Abstract:
You store data in SQL Server, but you don’t have enough work to keep a full-time DBA busy.
In just one session, you’ll learn the basics of performance troubleshooting, backup, index tuning, and security. Brent Ozar, recovering developer, will teach you the basic care and feeding of a Microsoft SQL Server 2005, 2008, 2012, or 2014 instance and give you scripts to keep you out of trouble.
Program Committee feedback:
- Accidental DBA’s are a big group and this session sounds like a good session form them.
- Great name and great description. Looks like a great session.
- Nice story. Clear objectives and tells me what I’ll learn. Then I get scripts woo-hoo! I’m there!
No negative comments, so good news there. I didn’t correct the spelling/grammar mistakes in the comments and I’m not going to make fun of them because I’m just happy to have comments at all, although I will point out that they ding US for spelling or grammar mistakes. (We’re all volunteers here, so no calling the kettle black.)
Conquer CXPACKET and Master MAXDOP in Ten Minutes (Lightning Talk, Accepted)
Abstract:
CXPACKET waits don’t mean you should set MAXDOP = 1. Microsoft Certified Master Brent Ozar will boil it all down and simplify CXPACKET to show you the real problem – and what you should do about it – in one quick 10-minute Lightning Talk.
Program Committee feedback:
- fits criteria, light talks no demos are ok, description looks good.
- Excellent introductory topic that is confusing even for seasoned DBAs. Topic appears well laid out and suited for a 10 minute lightning session.
Note the first comment – “light talks no demos are ok” – and read between the lines. The reviewer is saying that for a full session, he requires demos. This is where it gets to be fun, putting yourself in the reviewer’s head.
Watch Brent Tune Queries (Rejected)
Abstract:
Ever wonder how someone else does it? There’s no right way or wrong way, but in this session, you can peer over Brent’s shoulder (virtually) while he takes a few Stack Overflow queries, tries various techniques to make them faster, and shows how he measures the before-and-after results.
Program Committee feedback:
- Great topic and aimed at correct level I feel – should attract a big audience.
- I don’t really know what BE CREEPY is, and couldn’t find it on a search. I will also admit that the title and knowing who this is didn’t cloud my review (though I wanted to tweak the subjective rating even higher because of his involvement.), I like the idea of room sized mentoring going on and I expect the speaker rating will balance out my enthusiasm. I thought the abstract was the weakest part of the entry. I would like 5-10% slides for a session like this, to make it easier to remember later…
- I’ve seen this before. It’s great.
- avoid including names in abstract looks like some formatting issues. never heard of this guy, does he work for Microsoft?
- catchy abstract – interesting topic. clear goals. love the fact that its 100% demos.
About not putting names in the abstract – this isn’t bad feedback about me – it’s bad feedback about the process. Here’s why.
First, note that both of my accepted talks have my name in the abstract, and in both cases, the reviewers didn’t even mention it in the comments. That means this isn’t an abstract problem – this is a committee consistency problem.
Second, the title of the session is Watch Brent Tune Queries. I used to call it “Watch an Anonymous, Genderless Person Tune Queries” but the turnout just wasn’t as good.
Finally, when I was an attendee, I didn’t know most people by name. I did know a few major folks, and I wanted to make sure I didn’t miss their sessions. Call me vain, but given the way people seem to react to me at conferences, I appear to be A Big Deal. (It still makes me laugh when people call me a celebrity – until I can walk into e by Jose Andres without a reservation, I’m not a celebrity.)
So no, I’m not changing this approach, but I’m sharing it with you, dear reader, so that if you want to get a higher chance of getting into Summit, keep your name out of the abstract. And gender too. Certainly no sexual preferences, because they don’t want to hear anything about your individuality or background.
About 0% slides vs 10% slides – note that one reviewer loved 100% demos, the other said she wanted 5-10% slides. This is why Jeremiah says conferences end up like cheese pizza – you have to try to satisfy every single committee member, and that’s simply not going to happen. I bet if I’d have put 90% demos, the person who loved 100% demos would have been fine with it. (Technically, the session has 14 slides, but I rip through those in 90 seconds because they’re just takeaways for the reader to reference at home. 90 seconds out of 75 minutes is 2% slides, so I rounded to 100% demos.)
Cool Story, Bro: The DBAreactions Guide to SQL Server 2014 (Rejected)
Co-Submitters: Doug Lane and Jeremiah Peschka
Abstract excerpt (because PASS doesn’t show full abstracts after sessions are denied): “You’re hearing about the new features in SQL Server 2014, and you’re not quite sure where to start…”
Program Committee feedback:
- Goal 3 does not seem to have any purpose, therefore I consider there to be only two goals.
- The topic is probably the most popular one at the moment, and subjectively speaking this sounds like it would be a lot of fun. I knocked the abstract down a point mainly because of the goals. They are funny and entertaining, but strictly from the abstract perspective they do not give a clear idea for what the goals of this will be.
- Parts of abstract and goals do not seem to reflect a serious commitment to presenting a useful session. This topic will likely be covered in depth by Microsoft speakers given the timing of this years Summit.
Response:
In all seriousness (because like the Internet, the PASS Summit is serious business), we were pretty sure this would be rejected. Again, cheese pizza philosophy here – this session would draw a small group of raving fans, but it’s never going to appeal to a wide audience. That’s okay – was worth the shot.
500-Level Guide to Career Internals (rejected)
Abstract excerpt: “This is not yet another career session that tells you to be friendly and network. Forget that..”
Program Committee feedback:
- Name in the abstract must be avoided to ensure that the reviewer is not influenced by the speaker’s name. Session name is brilliant and well designed for the SUMMIT audience. Level 500 should be better explained in the pre-requisites as it’s considered as “level of skills to attend ” and not complexity for the speaker. However, given the speaker (supposed to be unknown) the session is deemed to be a success.
- Not sure that there’s anyone that would not want to attend this session.
- Contractions throughout. Individual experiences might not apply to everybody. Level 500 in utterly inappropriate.
- The title is not very clear. What is “career internals” anyway? Abstract is well written and makes clear what the speaker will present, however comes with some negative remarks about networking.
About the name – okay now this part is just flat out wrong. When I pick up a professional development book, the very first thing I check is the author’s credentials. If someone’s gonna tell me how to be successful, I want to know that they’ve been successful.
About the contractions – ah, that’s interesting. DANG, I used a contraction. See, I write like I speak. I gotta work on that.
About the 500 level and internals – I totally understand. Again, wrote this one with my oddball sense of humor angle, and I was pretty sure it would not (see, I didn’t use a contrac…oh, dammit) get in.
About the networking – if you want to scale, physical in-person networking is dead. Yep, I said it. It still works to sell stuff to old white guys in suits, but everybody else is online, and networking is very, very different here. (That reviewer is probably the one who most needed to attend the session, sadly.)
Power Tuner: DBAs are the New BI Performance Tool (pre-con, rejected)
Abstract excerpt: “How can you make SQL Server fast when you can’t predict the queries coming from SSRS, SSAS, SSIS,…”
The selection here is a little different – pre-cons are chosen by someone, we don’t know who, and we don’t know how. I’m actually fine with that because PASS has serious money on the line here. Pre-cons bring in big money. For example, I was told that our pre-con last year was the top-selling pre-con of all time, bringing in around 250 people. (That’s about $125K USD in revenue for PASS.) This year, not accepted, and I’m totally fine with that – PASS needs to do what’s right for them.
But I want to share the feedback anyway, because it’s valuable to other pre-con submitters:
- Precon addresses problems many users may face. SQL 2014 is a brand new version so this may entice some users to check out this session
- Thanks for the abstract, It’s good to have someone talking on DW solutions mainly on Performance tuning, the abstract topic is quite not match along with the abstract content. Also , prerequisites is not completed. Thanks
With only 2 comments, I’m not quite sure how to take this one, but I’m fine with it.
Life at Stack Overflow: My Developers are Smarter Than Your DBAs (Nick Craver, rejected)
Nick Craver (NickCraver.com – @Nick_Craver) is on the Site Reliability Engineering team at Stack Exchange, the company behind StackOverflow and DBA.StackExchange. He’s the main developer behind the open source SQL Server monitoring tool Opserver, and he knows more than I do.
I did volunteer to help with a few of his abstracts, though, which he’ll “thank” me for later because none were accepted. He volunteered to share his feedback here too though:
Abstract excerpt: “What happens when you don’t have a DBA getting in the way? Awesome, that’s what happens.”
Program Committee feedback:
- I think this would be good for dba’s \ devs and general interest from attendees. Obviously a high profile installation and in the current climate of cross-functional teams – very appropriate.
- So, on the one hand, this presentation concerns me with the snarky title. Get better DBAs! On the other hand, the abstract text is interesting, in that what it could be suggesting is that collaboration works (but dismissing the concept of DBA as a team member ruins this one for me). If you have data developers that know what they are doing, that is great. I see only mention of performance and nothing of data integrity, which is acceptable for certain sorts of data, but not others.
- This is very interesting. You can’t get too many opportunities to peek on this type of databases so this is a must see.
- Well written abstract that conveys business value with techie stuff. Reference to real-life examples are make this soud like a session that an attendee will want to attend and also possibly use to leverage mgmt for financial support of summit attendence.
- Session name is not very “nice”.
- Could be interesting
- this is a very unique and interesting topic. very catchy and well written abstract. clear goals – relevant and useful. lots of developers will be interested in attending this session
Not surprised there – the title was purposely over-the-top to get attention.
Note that there’s absolutely no technical feedback here – zero mention of title, abstract, prerequisite, level, demos, etc. Just purely personal feelings. To me, that’s a winning abstract – one that challenges people to think differently and provokes conversation. That’s not what wins in committees, though.
Failure is AlwaysAn Option: Keeping Stack Overflow Online and Fast (Nick Craver, rejected)
Abstract excerpt: “Stack Overflow’s Nick Craver will take you on a tour through their AlwaysOn Availability Groups…”
Program Committee feedback:
- I appreciate the real world example of this abstract. It is helpful to learn from the experiences of others.
- Not sure how this is a 200 level and the session prereqs are supposed to be taken for real? 25% demo and yet in additional notes says he will demo how stuff is working but I’m sure the % is about 50. No conclusion sentence on what I will learn.
- The prerequisite is inadequate to allow the attendee to determine if they have the requisite knowledge to understand the material.
That’s great feedback! Really helpful – I’ll pay more attention to prerequisites now.
Opserver: Keeping an Eye Your SQL Servers, For Free (Nick Craver, rejected)
Abstract excerpt: “You’re a DBA or developer who needs to figure out why your SQL Server is running hot, but you don’t…”
Program Committee feedback:
- Abstract could be written better. The topic has brilliant offering to beginner DBAs.
My guess, and this is only a guess, is that by accidentally a word in the title, Nick was totally doomed. No comments because people probably just low-voted right then and there and bailed. (I wish I had the whole abstract here to check though.)
What You Can Learn from Our Failures
The big takeaway: reviewers are independent human beings with totally different opinions on what matters in a session. If you want to get in, you have to appeal to as many as possible, which means making your abstract as inoffensive as possible. I’m still stunned and grateful that “Developers: Who Needs a DBA?” got in, and I’m not surprised that the rest didn’t.
I’m still going to keep taking chances, pushing boundaries, trying different session styles, and putting my name in my abstracts. For me, it’s about being the speaker I want to be. Some conferences will have me, and some probably won’t. Sooner or later, I’m going to get outright zero sessions at PASS, and that would be totally okay too – competition from other speakers is a GREAT thing.
After all, that’s why I just spent hours putting this post together – to help you beat me.
Because I want to see good sessions too!
13 Comments. Leave new
Brent,
This is excellent and an example of the kind of feedback potential speakers need to improve. For the record, I would have attended all of Nick’s sessions had they been accepted. How about having him on as a “guest star” under the Brent Ozar banner in the near future? It’s a given I’d attend your team’s session even from an overflow room. 😉
Ron – thanks! I agree about Nick, and we’ll be working on doing a webcast with him.
To be fair, I said “Conferences designed by committee are like cheese pizza – nobody gets what they want, but at least you get to eat.”
Brent,
It reminds me of a lot of school systems, the funny, not so serious teachers are frowned on by the school’s administration but they are also the ones that most students remember and learned the most from.
Chris – yep, exactly. My mad science biology teacher was the only fun thing about science in high school, hahaha.
What I’m taking away from the list of rejected sessions is that I wish the board would dare to be less boring (although certainly not all approved sessions are boring!). You have to risk offending someone to be interesting, but it’s worth it.
Is there an Unboring Conference for databases?
James – I’m totally not an expert on this, so take this with a grain of salt.
I think whenever you add more people to a process, their decisions become more average and less extreme (except when groupthink kicks in.) As long as there’s groups of people making the conference selections, this is going to happen. Even if you guided them to be more open to unusual sessions, each individual member of the group will tend to pick different unusual sessions. If it takes votes to get in, the cheese pizza effect will still mean the mainstream sessions get accepted because there’s a limited number of voters.
The way around this is to dramatically increase the number of people with a word in the selection process – meaning, voting by the attendees.
PASS is lucky enough to have a large existing attendee base. They have email addresses for past Summit attendees, which are the most likely customers to understand what makes a good session, and be the types of people who have budget money to return to the Summit again. By asking them which sessions they’d like to attend out of the list, PASS would get a better picture of which niche sessions would actually attract people. Specifically, the most valuable people – ticket buyers.
SQLbits is another SQL Server conference that uses a modified version of this approach. They invite the public to vote on which sessions they’d like to attend. Votes alone don’t dictate which sessions get picked – they’re only a part of the selection process – but it’s very successful, and it’s my favorite of all the SQL Server conferences out there right now. I happen to be flying over to the UK to attend next week!
Hi Brent, yes, I think some form of attendee voting would probably make PASS more interesting. Not sure if the cheese pizza effect is real, but certainly it’s plausible enough. Anyhow, glad you are coming to SQLbits. One of my coworkers will be in your precon.
James – thanks! The cheese pizza effect definitely is real – there’s been plenty of books about it, especially in regards to product design by committee.
So, based on everything I’ve read here, I need to definitely avoid “Babycakes Clary’s Guide to Eating Cheese Pizza Like a DBA” — unless I include the proper prerequisites.
HAHAHAHA, nice.
[…] improve the process, and many others have offered some suggestions (Adam Machanic, Andy Leonard, Brent Ozar, Brian Kelley, Tim and Lori Edwards, Jen and Sean McCown, and probably more, sorry if I missed […]
[…] 2014 feedback about my Summit abstract submissions […]