Unless you take a close interest in the politics of PASS you may have missed the kerfuffle around it changing its name from the “Professional Association of SQL Server” to just PASS a month or so ago. Blog posts from Andy Warren and R. Brian Kelley seem to sum up a lot of people’s feelings. There was a similar outcry on Twitter following an announcement about the way speakers will be selected for next year’s PASS BA Conference. In both cases I read the news when it came out and didn’t give much thought to it, because I agreed with what PASS were doing and I could understand why they were doing it. The strength of feeling these moves stirred up surprised me, and it made me realise that what made perfect sense to me was not so clear-cut to other people in the SQL Server community and needed to be defended – hence this blog post.
Let me start with the name change. I don’t think PASS will ever be anything but a SQL Server-centric organisation, whatever its name. However the definition of what is “SQL Server” and what isn’t has blurred in recent years. You could argue that PASS should confine itself to products that have “SQL Server” in its name, but if you were to apply that rule strictly you’d exclude Azure SQL Database. Aha, you say, that would be silly because the technology behind Azure SQL Database is so closely related to on-prem SQL Server and people who use the former will almost certainly use the latter. But in the same way Power Pivot is SQL Server Analysis Services running inside Excel – it’s more or less the same code base, the development experience is the same, the development team is the same, and the DAX language is identical in both SSAS Tabular and Power Pivot. The SSRS team gave birth to Power View and a lot of MS people closely associated with SSIS now work on Power Query. The worlds of corporate BI and self-service BI are drawing ever closer and soon it will be difficult to say where one starts and the other ends. A lot of SQL Server BI consultants and companies (like me) are now using and building Power BI solutions in addition to the SQL Server BI stack. This is why PASS is right to concern itself with Power BI and self-service BI: it’s a big deal for a lot of PASS community members.
What’s more, Microsoft is working on a number of new product that may have no technological link to SQL Server but will be used and championed by members of the PASS community. Azure Document DB and Azure Machine Learning are two good examples. To me it’s right that PASS takes an interest in these products even if other technical communities do so as well. Anyone with a background in SQL Server – anyone with a background in managing and analysing data with Microsoft’s traditional data platform – will have a natural interest in Microsoft’s new tools for managing and analysing data.
The argument against what’s happening seems to be that it means PASS is no longer focused solely on SQL Server, that it will end up spreading itself too thin. The people making this argument, as far as I can see, are from the DBA side of the SQL Family who equate “SQL Server” with the SQL Server relational database. That’s an easy assumption to make for those who only work with the relational engine. It’s true that “SQL Server” is the name of the relational engine, but “SQL Server” has also meant more than the relational engine for a long time: SQL Server is a suite of tools, and the relational engine is just one part of it. I’ve been a SQL Server MVP for nine years and have almost nothing to do with the relational engine. I started working with OLAP Services with SQL Server 7 and spoke at my first PASS event at (I think) the 2005 PASS European Conference in Munich. I haven’t been involved in the SQL Server community as long as some people, I know, but I feel like I’ve been part of it for long enough for my definition of what “SQL Server” is to carry some weight. As I’ve just argued, SQL Server Analysis Services, the tool that I have worked with for almost all of my professional career, has found a new expression in Power Pivot (even while traditional SSAS still exists) and so I think it is right for PASS to follow me as my career changes. This isn’t PASS losing its focus; this is PASS keeping its focus on what SQL Server has become and what its members are working with. I’m not leaving the world of SQL Server behind. SQL Server itself is changing and I’m changing with it.
Nor do I see any evidence of this stopping PASS from providing support and education for DBAs and those who work exclusively with the relational engine. As the PASS community has grown, as the PASS Summit adds yet more tracks, as other events like SQLBits and SQL Saturdays have taken off, I see that the economies of scale that come from the wider definition of “SQL Server” to include the relational engine and the BI tools have provided even more opportunities for those who are focused on the relational engine alone to learn and grow. Each sub-community may have a smaller slice of the pie, but the pie itself is much larger than it would be if we all went our separate ways.
This brings me on to the topic of the BA Conference. Andy Warren’s assertion, in the blog post I linked to above, that the BA Conference should have been the BI Conference is an opinion I hear repeated a lot and one I disagree with strongly. Before I do get onto that topic, though, I want to be clear that I’m not picking on Andy in particular – I’m engaging with him because he puts his side of the argument in a clear, thoughtful way which I respect (this more recent post is a great example of why I admire him and his thoughts on PASS). Unfortunately this particular discussion risks being turned into a very unpleasant argument, and I hope the PASS Board takes action to stop this happening. Twitter is where this is most obvious: it’s easy to be angry in 140 characters, less easy to have a nuanced debate. The tradition of heckling by certain members of the DBA community on Twitter during PASS Summit keynotes while BI features are announced is something that particularly p*sses me off: I feel like it’s rude and disrespectful to the large number of people in the audience who are interested in these features, and to the people on stage announcing them. Let’s stop this kind of thing, it can only be divisive.
OK, back to the topic of the BA Conference. When it was first announced I wrote a post entitled ”The PASS Business Analytics Conference is not the PASS Business Intelligence Conference” but I think my arguments here need to be clearer. First of all, I do not agree that BI content should be split off from the PASS Summit into its own conference. Some DBAs are not interested in BI; some BI pros like me are not interested in relational engine topics. However we are in the minority. I know from my friends and customers that a lot of people who work in the SQL Server community now deal with both the relational engine and some or all of the BI tools, and having the PASS Summit cover both BI and DBA topics at the same conference makes it appeal to a much wider audience than it would if it concentrated on just one or the other. Indeed, part of the reason why I felt disappointed with this year’s PASS BA Conference, and why I felt it was a bit of a failure, was because it had fallen into the trap of being a PASS BI Conference.
So why then should PASS have a BA Conference, and why is BA<>BI? Let me give you an example. Last week I ran a private SSAS Tabular/DAX training course for a large company in London. The people on the course were traditional corporate BI developers, smart guys, the kind of people I have worked with all my career. This week I’m going back to the same company to deliver training on Power Pivot, DAX and Power BI to the business analytics team of the same company. A lot of technical topics I’ll be covering will be the same but the people attending will be very different. They are equally smart guys who work outside the IT department, but in close association with the IT department, analysing data to answer business questions. They don’t have the kind of technical background that someone in the IT department might have but they do use many of the same tools and technologies (SSAS as a user; Power Pivot and Power BI; Excel). They ask the business questions and they then work together with the IT guys to find the answer. This is the BA community.
A BA community has always existed within the Microsoft ecosystem, originally around Excel and Access, but with the advent of Power BI and the wider self-service BI and Big Data movements it has grown in size and confidence recently. Evidence of this growth includes the “New Excel” user groups that Rob Collie is involved in organising, and the London Business Analytics User Group that Mark Wilcock organises. Should PASS reach out to this community? Yes, I think so. There is a clear overlap between the technologies that the BI side of the PASS community uses and those that the Microsoft BA community is using. Also, I think the BA community can benefit from the passion, experience and infrastructure built up by PASS. As far as I can see, the BA community is also happy to collaborate and engage with the PASS BI community too. In the past few years I’ve made a lot of new friends in the Excel world, for example Excel MVPs like Bob Phillips, who have this attitude. BI pros and BA pros are now working side by side in the office, so it makes sense that we come together as a professional association.
The BA community is not, however, interested in being swallowed up by an organisation that identifies itself purely with the IT department. It is interested in joining an organisation that has IT focused people as a major component but which also recognises the importance of collaboration between IT and the business to work with data. This is why the PASS name change, and especially a separate BA conference, is important. DBAs will only ever want to attend the PASS Summit. SQL Server generalists, and BI pros, will also want to attend the PASS Summit. Some BI pros will want to attend the BA Conference as well as or maybe instead of the PASS Summit. BA pros will only want to attend the BA Conference. BI pros like me will be presenting on topics like Power BI alongside Excel pros at the BA Conference; this is where the overlap between the existing BI community inside PASS and the new BA community is obvious. There are also a lot of topics such as machine learning, R (which is a big part of Azure Machine Learning as well as a massive topic in its own right), Hadoop and pure Excel that do belong at the BA Conference and do not belong at the PASS Summit.
As I already said, I felt like this year’s second BA Conference was a bit of a step back from the first because it had too much pure BI content. Part of the reason for this was, I suspect, because PASS relied too heavily on its existing speaker community and did not make enough effort reach out to the BA community’s speakers. I believe the PASS Board has come to the same conclusion as me and this is why the BA conference next year will not have an open call for speakers. Brent Ozar did a valiant job of trying to imagine why this has taken place here; he falls a little wide of the mark with his explanation in my opinion, even if he’s dead right that the PASS Board need to make a clear and positive case for why they are doing this. Here’s my take. A large Microsoft BA community undoubtedly exists. There are a number of great speakers out there too, but they are not yet convinced that PASS is serious about engaging with the BA community so they are cautious about investing the time and money to speak at the BA conference. What’s more the economics of the BA community are very different from the economics of the traditional PASS community. In the world of SQL Server there is a virtuous circle that allows someone like me (or Brent, and any number of other well-known speakers and trainers and consultants) to invest our own time and money in attending the PASS Summit, SQLBits and SQL Saturdays and so on because we know that it strengthens the community and therefore benefits our own businesses, which are based on high-rate consultancy and training. I’m not saying that we do this cynically – we are genuinely motivated by a love of the community, it’s just that what benefits the wider community benefits us as well. Unfortunately this virtuous circle does not exist (yet) in the BA community. Potential speakers in the BA community are more likely to be full-time employees, or academics, or people like Mr Excel who make a living from selling their content to a mass market, rather than from a small number of customers who pay high rates. They may not be able to afford the time and money to come and speak, and giving away their content for free at the BA Conference might be detrimental to their business. If PASS is going to make the BA Conference work then it will have to approach these potential speakers individually, convince them that the conference is worth their while to attend, and probably pay travel expenses and maybe also pay them a rate. Let’s not forget that this is normal practice in many other technical communities, and I know that top speakers have in the past been paid quite handsomely to speak at TechEd.
The last thing I want to say is that it could be argued that PASS is changing its name and focus because Microsoft wants it to, and indeed that it depends on Microsoft so much that it has to do Microsoft’s bidding. I don’t accept this argument though. PASS should support the interests of its members regardless of the whims of the marketing folks at Microsoft. My point is that I believe that the changes that PASS has made have been for the right reasons, because they are in the interests of a large number of people like me from the BI pro community who have been part of PASS for a long time. The forces that Microsoft is responding to in the wider world of data are the ones that PASS itself must respond to.
I think this post has gone on for quite long enough, and if you’ve read this far then I’m very grateful. I expect that I’ll provoke yet more debate by what I’ve said here, but I also think it needed to be said because the arguments so far have been one-sided. I don’t pretend to speak for the PASS Board or have any more inside information than anyone else already involved in this discussion. However I do believe that the more we talk about these issues, the more likely we are to understand each other’s points of view and come to some kind of agreement that we are all happy with.
Chris, thanks for the long post – it has been very helpful to me in order to understand what’s going on in PASS BA conference. I read the announcement, but I didn’t thought to the different BA communities and I agree what is the effort that is going on.
I constantly wonder why other technologies and products don’t have a community so strong as PASS. Maybe SharePoint can be compared to, even if there isn’t a “single” organization grouping all the communities (at least to my knowledge). There are many explanations, but I don’t see as an issue the fact that PASS is extending outside its original boundaries, covering “all about data” (my personal definition).
I am not sure that the PASS BA conference operation will work as expected, at least in terms of “community building”. But it’s clearly something to try, considering that I agree with you that this year it wasn’t a step forward.
However, it’s not only about speakers, it’s also about content. I’ve seen sessions this year that had a very low number of attendees, despite the topics covered were in the “buzzword” area.
Interesting post!
Just to make sure I understand, you’re saying that the BA users wouldn’t go for an IT-focused organization, and they use different products with some overlap with PASS, and PASS’s presenters couldn’t give them the right material.
So why is PASS the organization to run their conference?
Replace “analytics” with “Oracle” and your entire post still has the same point. Many SQL Server professionals also interact with Oracle, but Oracle users won’t go to a conference run by PASS, and if PASS wanted to run a legit Oracle conference, they’d provavly have trouble building a full robust agenda using PASS speakers.
Reading your post, I think you’re driving the point even further home.
No, I think you’ve missed the point. A lot of MS BI pros now work with Tableau and Qlik, but I’m not advocating that PASS tries to cover those products any more than it should cover Oracle. The BA community is a new community of users for Microsoft BI tools, one that is growing, one that wants to engage with the organisation that represents the older Microsoft BI community -PASS. It has different needs to the existing MS BI community, ones that PASS will have to work out how to cater for, but ones that the existing BI community cannot afford to ignore if it is to succeed in the long term. So PASS is not trying to branch out into areas that it shouldn’t be meddling with, it is dealing with changes within the existing MS BI community.
Chris – no, I think that’s *exactly* the point. I’ve heard other BA/BI professionals tell me that the BA conference is going to cover non-Microsoft technologies. I don’t see any vision whatsoever about where this thing is going.
You’re right, the focus needs to be clearer, but I think it’s pretty certain that the scope will be limited to Microsoft technologies. The confusion probably stems from the topics such as R: it’s a non-Microsoft technology, but if you’re going to do anything useful in Azure Machine Learning then you have to use R.
The “elephant in the room” is Hadoop though; you make the point that the BA content is aimed at the people without a technical background who are using Excel, R, etc. etc. to answer business questions; where does material on SQOOP (for instance) fit into this continuum? The delineation between BA and BI seems to hinge more on what’s the technology *du jour* than on any attempt to define the audience.
No, I’m not saying that all of the content of the BA conference should be aimed at people with a non-technical background. A lot of it will be, but traditional BI pros and data scientists will also be attending and they will want technical content. As well as Hadoop there are other new MS products such as Azure Machine Learning that will be used by BI pros, whose output will be consumed by the BA community (who need to understand how this stuff works, and who often have a stronger background in maths and data analysis than IT guys which means they are capable of understanding it) which are probably well outside the scope of what the Summit (which should retain it’s existing focus on SQL Server) should cover. These topics fit naturally into the scope of the BA conference.
Chris – you touch on a great point, actually. What *is* the scope of the BA conference?
Your reply mentions technical and non-technical users, Microsoft and non-Microsoft tools. It’s starting to sound like the Zombo.com of conferences, heh. That’s not your fault – there does have to be a vision for this, and the vision has to be clearly communicated to the paying customers.
The vision, as I understand it, is that it’s the conference for everyone doing data analytics using Microsoft tools. But I agree, it should be clearer. Instead of “Hadoop” PASS should really be saying “HDInsight” because, let’s face it, that’s what we’ll be talking about.
So, the material on “Hadoop for the SQL Server Professional”. Would you say this was intended for:
* Traditional BI Pros
* Data Scientists
* People with a non-technical background
* The BA community
* Another group not yet enumerated in this comment thread
Further, it’s a bit disingenous to point out that the output of Azure ML will be consumed by this nebulously defined “BA Community”; the BA community (and indeed the BI community, whoever they are) will also consume the output of SQL Server itself, the performance of which will be largely be determined by database design and indexing – and of course hardware – none of which appear to have featured in this year’s BA conference.
Applying Occam’s razor, the simplest conclusion to be drawn is that the content of the event is being driven by the marketing objectives of the principal sponsor.
I would say that it was aimed at traditional BI pros – the SQL Server professionals of the title. I know plenty of Microsoft BI pros who have been asked to turn their hand to HDInsight in the last year or so, Jamie Thomson for instance (see http://sqlblog.com/blogs/jamie_thomson/archive/2014/09/03/updating-hadoop-user-password-for-hdinsight-emulator.aspx and other posts).
Re Azure ML, the output needs to be consumed by and understood by the BA community. These guys understand a relational database and know what a table of data is; they may not understand what Azure ML is doing, or how it produces its results, or how these results should be interpreted. That’s why they want to learn about it, even if they’ll never use it themselves.
Your post and replies focus primarily on Microsoft-based data solutions for PASS. When I read the announcement and the direction, I don’t get that at all. I get a sense they are trying to be like The Data Warehouse Institute and be platform agnostic.
Which is why I wrote that I think there is still a great need for a SQL Server specific org. To your point about some saying SQL Server meaning just the relational engine, I don’t. I mean the whole suite, which is pretty sweet. (Sorry, couldn’t resist the pun.)
Thanks for taking the time to write this. I guess I see lots of different sides to the argument and don’t necessarily believe there is a single correct point of view with respect to some of the changes that have happened and other things that *will* happen. Like it or not, the industry is changing and organizations like PASS do have to change with the times. This is why I decided that SQLSaturday Cambridge ’12 would include dedicated BI, Azure and SharePoint tracks alongside the more traditional DBA and DEV topics. The next one will be even more diverse. As such I think PASS have quite rightly recognised that too, but I do wonder if sufficient thought has actually gone into the name change (I don’t have too much insight into how much time was spent on this so I could be wildly off base). For instance PASS was already covering much of these subject areas and have done so successfully for a long time, so one thing I fail to understand is why the name change (as it stands) is really going to make a whole jot of difference to this. For instance were BA/BI professionals really turned off by the name as it stood and will the new name make a difference to their perception of the organisation now? I don’t think so to both questions. With respect to the change to the submission of sessions to the BA Conf, I have absolutely no problems with the perceived reasons for the change this time, but I do think it is a little short sighted for us to think that a complete block on community submissions was necessary so that PASS can instead approach only those they want. It was always going to have a negative reaction by some of those in the community who would want to submit, so just for that reason alone perhaps it wasn’t entirely the best solution? I’ve read the arguments on this and fail to understand why PASS couldn’t approach who they wanted after taking into account of the community sessions. They can then pick the best of both worlds. The traditional PASS Programme Committee approach does not need to apply. Selection by committee can still be a closer to a dictatorship (i.e. a very small committee) -and as you know this works pretty well for SQLBits. I do agree with Brent on his argument about the Oracle angle (i.e. where do we define the boundaries), but I would also qualify that by saying that PASS does even try to cover Oracle integration and use for SQL Server professionals to a smaller degree (disclaimer:I have recently taken over the OracleVC). In summary whilst I don’t believe that sufficient thought has gone into some of these recent changes I do believe that change is necessary. The BA Conference does need to prove that it has a strong market and I sincerely hope it will. I hope the changes that have been made will help steer it in that direction, but if they don’t then serious thought and discussion is necessary about its (and PASS’s) future direction. Either way I will continue to support and promote PASS (and its decisions) as best as I can, but I hope they do not lose sight of the fact that they exist to support the Community so that we can do the same to them.
Mark (Retracement) – yeah, exactly. There’s an Oracle virtual chapter that works just fine without requiring a name change to PASS.
You bring up a great point about the BA Conference needing to prove that it has a strong market. Why not let it run as a virtual chapter and start regional grass roots events first before spending money on a national conference and paying speakers to do it? Just as SQL Server user groups started at the grass roots level first before proving their need for a national conference, there’s no reason we couldn’t try starting smaller BA events first rather than putting hundreds of thousands of dollars on the line here again.
Brent, I can see top down working, as much as I prefer slow growth bottom up. The big concern I have is that when it comes to marketing, without a big list and solid social presense they are just selling to the “old” PASS. Well, that’s one of many concerns!
Andy – exactly. Top down costs much more money (for push marketing as opposed to inbound) and they don’t have a target. I’ve repeatedly asked how we’re going to get the word out to prospective BA attendees, and I haven’t heard an answer yet that doesn’t involve truckloads of marketing money.
Brent, on top of that, it doesn’t feel like the marketing strategy for the Summit is sophisticated (maybe I’m wrong, marketing guy I’m not). Just so different to start something new vs established. Price bumps aren’t going to be enough.
I cannot disagree about anything you have said tbh (shock horror!) and this is very much how the situation appears to be. I know there are several smaller PASS BI Events and a couple of new smaller BA Events, but most of these are in their infancy and from what I have seen there certainly doesn’t seem to be a stampede to them (tho again I could and hope I am wrong). I think Chris is right about some of the groups he mentioned, because I had a look at some last week at was surprised how big their membership was, but then does it automatically follow that these are the people who this event will attract? I’ve seen other groups such as IoT related ones and their memberships are massive (how does this translate in attendees I wonder?) – but should PASS also aim to target them and those technologies also. I think this is the biggest problem with the current name change is that it seems that we have also lost our identity.
I feel particularly connected to this issue because I was not entirely convinced about the conference when it was first announced those couple of years ago and always saw it as SQLRally’s replacement, although PASS always denied that fact. And in fairness I have heard some of their arguments for it’s cancellation and I suppose some of them are fair, although I don’t entirely agree with all of them. But I do have rose tinted specs about those events having been (I think) the only European to have spoken at both. Orlando and Dallas will always have special memories for me and I am still in mourning! Anyway I digress, back when the BA Conf was announced, it felt like it was an event that was trying to find an audience and despite the strong optimistic comments from the BA/BI Illuminati at the time I don’t think it has really delivered. That doesn’t mean this one won’t though -I sincerely hope it does.
Any half decent Poker player knows that you should always try to play the odds, but if you are going to go all-in without a good hand then you had not only be very good at reading your competition but also hope to heck that they actually do what you think they will. I don’t think I’ve heard a compelling argument either why PASS are playing this hand given what we currently know, but I hope they are better at reading the competition than I am because I seriously don’t want them to fail.
>>Why not let it run as a virtual chapter
Brent, there has been a PASS BA VC since February 2013. And as I said in my post there are already grass roots user groups.
Chris – exactly, and do they have a higher level of turnout than other subjects? If not, why focus on this particular one?
I think the Board – and I agree with them here – believe that there’s the greatest potential for growth in the BA area. If PASS has only got the time and energy to focus on one area to expand into I think this one is as good as any.
(I should note – I know full well what the answer is, and I’ve got a blog post going live on that tomorrow. Elect a bunch of BI consultants to the BoD, and the result is fairly predictable.)
Yes, unlike in other elections where I’ve voted, I’m certainly getting exactly what I voted for here.
Hahaha, yeah! I can see why the BI/BA crowd is really happy with what’s going on, and I just think the DBA/dev crowd doesn’t understand where the focus is coming from. If people wanted different results, they should have voted a different set of Board members in.
Chris, appreciate you taking time to post your view. I don’t think it’s nearly as simple as DBA’s and the SQL engine and I think that there is always going to be some group that doesn’t fit under the tent. I’ve been waiting to write something deeper about it until my head was clear, maybe this week:-)
I’ve heard of PASS but hadn’t really checked it out until this post. I am in the new BA segment. I have no interest learning about traditional DBA and RDBMS topics. My local PASS chapter seems to be fixed on those, but by contrast, almost all of the BA Virtual Chapter’s topics are of direct interest to me. Every business user I talk to has data challenges, and when they see Microsoft’s BI stack the first response is usually “How do I get that?”. We will need education as this takes off and it is good to see PASS is adjusting its conferences to meet this need.
Every proposal that I write these days has to speak to self service data analysis by the business, at least in the first instance, without moderation by the IT function which is generally seen as a bottleneck and impediment to business growth.
Typically time and money is wasted in organisations getting to the right data/information due to the Chinese whispers between business user, analyst, developer, analyst, business user.
I support PASS trying to get educated users and analysts into the tent such that requirements are informed by knowledge of the virtuous circle available between for instance Excel Power Tools and Tabular Analysis Services that can productionise tested end user models without unnecessary re-engineering.
Rather that than have to review an unscalable “proof of concept” that the organisation now relies upon that was “developed” in a trial version of QlikView or Tableau by a frustrated business user.