Are we too far gone by now for anything to be able to put us back on the right course?
52 replies (most recent on top)
There is no turning back.
The grassroots appeal to hardcore scientists was what made it into a big success. That generation of folks has since retired or moved on. The younger crops of hardcore scientists are not interested in paying for software that they can write using open source libraries. SAS reached it's apogee long ago, and is rapidly falling back to earth.
Denial is a real phenomena, and it's tightly woven into the cultural fabric of SAS. There is no pulling those threads, and no shaking those folks back to reality.
Why does this thread repeat after every layoff event? Denial, perhaps?
I was so hopeful when the previous CTO was promoted. As brittle as he was and as difficult as he was to deal with (they called him a prima donna in BIRD), he was one of the few people at SAS who could probably hold all of SAS 9 and Viya in his head, including the build pipeline. He was our best hope of having an actual architect at the head of R&D who knew how all the pieces fit together.
But then Dr. Goodnight promoted him to COO, maybe at his request or with his agreement, and any hope that Viya would have an actual architect went out the window.
Then I was hopeful when the current CTO was promoted. He said all the right things, and coming from DevOps he knew what kinds of problems we had to solve.
But as soon as he was promoted, he gave all of his problems to his subordinates, who refused to do anything about them because solving his problems didn't serve their agendas.
I'm not hopeful that a change in management will accomplish anything. Remember when the previous CTO was having meetings with his direct reports' direct reports (I think he called them "skip-level meetings")? He did that because his direct reports weren't telling him what problems they were having, not wanting to jeopardize their position at SAS. In other words, they were actively lying to him.
And that's always been the problem at SAS: every layer of management exists to conceal problems from the layer above, so they don't get blamed for problems that are probably not preventable and demoted, or worse either invited to resign or escorted off campus as the result of some "HR action".
100% agree, @aqu+1nHNoFH4
"But if you don't see a big management change, don't expect anything else to change."
Well, at least those garden level variety power players can now feel powerful, like they are "somebody" finally. There must be something about the view from the bridge of the Titanic that's appealing.
It's not true that "no one wants to hear it". Some of us agree with you.
"We're going to lose senior people, sure, but they'll be replaced by younger people who already have cloud skills"
This quote illustrates the problem. SAS actually has managers of such caliber that they believe losing experienced people is desirable -- even though, after losing many such people, SAS revenues have declined >20% over the past 10 years.
SAS can change, the same way Microsoft did -- by replacing its Ballmers with Nadellas.
But if you don't see a big management change, don't expect anything else to change.
Before 2021 there were ~100 highly skilled R&D veterans + younger Principal-level luminaries who worked hard to change the game … REAL Engineers. This group was a force multiplier and each “punched well above their weight”. Many were among the best mentors, tactical leaders and developers SAS ever had. Virtually all were being underpaid by $50-100K annually. Some even left for 2X what SAS was paying them.
Meanwhile the political operatives, retired-in-place and loyal yet under-developed rank-in-file stayed — though many in this group have departed due to packages, layoffs, or in some cases for other opportunities.
There are still a lot of good people at SAS but their numbers have been greatly diluted and many must be disillusioned.
The cumulative results of these factors are unfolding before our eyes.
No. SAS is a dying company. It will go IPO, the board will set targets for revenue and growth which the company will fail to meet even with aggressive cuts, and then it will lumber along headless for a few years, until ultimately its IP and core technologies will be sold to a competitor for pennies on the dollar.
Remember a few years ago when Dr. Goodnight said: "I think it's time to put R in its place"? How's that going? SAS is not winning the battle against open source. In fact, every company using SAS is trying to find a way to minimize its SAS spend. Is it their fault that Viya is not a replacement for SAS 9? No, it is not. Management has consistently failed to make parity with SAS 9 a priority, giving customers as many reasons not to migrate as there are feature or functional differences between the two.
Remember when the previous CTO announced the various client frameworks for interacting with CAS, e.g., Python and R right at the beginning of Viya? That was a great idea. How's that going? I'm confident that if we were to compare the total volume of code written in SAS for SAS 9 to the volume of code written in Python or R for Viya, there would be a difference of several orders of magnitude between the two, favoring the former. Yes, SAS 9 has been around much longer, but what does that mean? Well, even after removing the learning curve that is the SAS language (and macro, which is inspired), SAS's customers don't want to use open source frameworks and languages to interact with SAS, and they're not willing to pay to migrate their existing workflows to Viya. Viya is ten years old at this point, more than enough time for any company wanting to migrate to have done so several times over.
Remember when the plan was to merge SAS 9 and Viya at some point (before the pandemic)? How's that going? Not only was SAS 9 never merged with Viya, but now the company, including the remaining developers and testers and lumbering internal technologies like the CI / CD pipeline, has to support not only the legacy build / deployment technology (SDW) for SAS 9, but the build / deployment technology (Ansible) for Viya 3, and the build / deployment technology (Kubernetes) for Viya 4. Instead of one consolidated set of offerings, SAS now has three! And SAS 9 is responsible for the overwhelming majority of SAS's revenue. Viya only accounts for, maybe, 15 or 20 percent of SAS revenue. We'll only know for sure after SAS goes public and has to disclose its financials to its investors.
Seriously, SAS in the current era is bad at everything but analytics, and is in the same position that Microsoft was in at the end of the Ballmer era. Microsoft was unable to extend its monopoly on the desktop into mobile because the cell phone manufacturers saw it coming and didn't want Microsoft to commoditize the price of phone hardware the way it had desktops and laptops. SAS is being locked out of new markets for the same reason (and a few others).
SAS should have outsourced its CI / CD pipeline to an "aaS" company that knows what it's doing. More broadly, SAS should have maintained focus on its core competencies and not gotten distracted. SAS should have invested in cloud training and education. SAS should have valued the employees who were actively contributing to solutions to its problems in lieu of their own career advancement, especially in middle management. But it did none of these things. It actively drove those people away. Now they're working for Salesforce, SingleStore, NVIDIA, etc.
I actually heard one senior director at SAS explain it all away as (paraphrasing, since I don't remember his exact words) "we're going to lose senior people, sure, but they'll be replaced by younger people who already have cloud skills". Imagine that. His "big plan" was to replace high-paid veterans with low-paid college graduates that had taken three semester hours in "cloud" at NC State so that the company would gain critical skills by employee attrition.
I could go on, but why? I've said this all before, in conference rooms in buildings R and S. No one wanted to hear it then, and no one wants to hear it now.
I think that yes we can be on the right course. That course might be different from our current direction, or it might mean with a leaner and more efficient company. We are still populated with talented and motivated people who want the company to succeed and are willing to do the work. With the right leadership and appropriate staff, the company has a profitable future.
I think the company is being prepared to be sold, and I base this opinion on the failed Broadcom acquisition. And when it is sold, I am inclined to believe my line of work would be outsourced, so I am looking at other opportunities. For the 25+ years that I've been at SAS, it has been great company. But the last 10+ years, have been very challenging. I always come back to SAS dropping the ball with their university outreach. What a horrendous rookie like mistake.
There are too many competing factors and the world has moved on. Most successful institutions have a limited life span.
I think you’re goners.
I’m genuinely interested to hear what people think about this.