3 ms·
I just left Pivotal Labs myself, albeit a bit more voluntarily (after seven years), and Jeremiah isn't completely wrong, but he seems to think he's seen a huge
by ohwaitnvm 12y ago
I just left Pivotal Labs myself, albeit a bit more voluntarily (after seven years), and Jeremiah isn't completely wrong, but he seems to think he's seen a huge change over his three months working there in a role that is hardly one of Pivotal's core competencies.
From a former Labs perspective, Pivotal can really be thought of as several distinct groups working within the same office spaces. There is the one that the audience here is most likely to know; Pivotal Tracker. Then there is the consultancy, Pivotal Labs. Finally, there is the overall Pivotal organization (including elements of VMWare, EMC, Greenplum, and other EMC properties) focused on creating CloudFoundry.
Pivotal Tracker was, for a long time, written by consulting Pivots during time unallocated to any specific client. More recently, Pivotal Labs realized the value of the product, and divided off an entire office (as well as some support from other offices) to work solely on Tracker. For the closest thing to a startup experience within Pivotal (as opposed to consulting), working for Tracker is where to be.
The consulting practice has always been the core of the Labs business. Every office (save the Tracker HQ) has some amount of consulting Pivots, and any new office is established with 66%+ consulting personnel with the intent to gain traction within the startup market of the new city. Pivotal has _always_ consulted for both startups and enterprise level clients, and I take issue with Jeremiah's assertion that Pivotal has lost its way with regard to types of client engagement. He simply has no idea. As recently as Dec-Feb I was leading a team of a brand new startup through a Rails+Backbone rewrite of their existing PHP/Codeigniter app that they had decided they had outgrown; my engagements directly before and after that were both for established companies (enterprise, even) that needed specific software expertise and/or assistance with process within their corporate structure.
The final group is those devoted to CloudFoundry. The interesting thing here is that the developers and project managers (and to some extent the designers, though CloudFoundry needs very little visual design relative to its backend complexity - think Heroku) come as a mix of dedicated employees OF CloudFoundry, as well as support from consulting Pivots who are between projects or have a specific skill that can contribute to progress of CF. Consulting Pivots are told that they always have the option to rotate back out to non-CF projects, though there can be up to a couple of months of waiting while a suitable project or suitable replacement can be found. Let's be honest, EMC has given the directive from the top that CF must be delivered sooner rather than later, and that it is more important to EMC and Pivotal (not Labs, just Pivotal) that CF is delivered than that Labs engage yet another client.
So let's talk about Jeremiah:
Jeremiah was in marketing, which means he was almost certainly part of the CloudFoundry section of the company (the other branches barely know what marketing is), which has always been interested in delivering their service for use with startups, but in my opinion, is still at least 6-9 months from having enough stability and enough service integrations to be useful for anything more than a demo environment. I haven't tried deploying to CF in at least 4 months, so YMMV, and it's likely a lot more stable and well-rounded than I remember. Jeremiah talks about how he "stirred the pot on the distinction between corporate and startup marketing", which sounds to me like he was really much more interested in the consulting branch of the company and the startups involved therein, not the cloud services part. Consulting pivots probably make up 70% of each Pivotal Labs office, on the average, though there are many offices for EMC and Pivotal that contain no consultants from Labs at all. I remember being at a CloudFoundry all-hands meeting over a year ago wherein the number two director of CF specifically said that startups and the community at large around CF is secondary (or worse) to delivering an enterprise-ready solution for some rather large, well-known interested parties. Pivotal is exactly the three-letter acronym that Jeremiah "[does] not trust", but it has never not been. Pivotal Labs may share a space and employees with Pivotal, but it remains a noticeably distinct entity. With EMC's purchase of Labs and subsequent spinoff of Pivotal, things became increasingly more corporate, but that has been true since March of 2012, a good two years before the glory days that Jeremiah so fondly remembers.
Now for me, why did I leave Pivotal? I was getting tired of working on the same early-stage startup code every other project, punctuated by rescue missions for larger (Fortune500) companies who had no interest in the process and discipline practices that Pivotal can teach (pairing, tdd, stable/reasonable hours, tracker stories) and really just wanted us to shut up and fix their damn code. Pivotal Labs has been taking more clients with mobile development needs as well, and my disinterest in coding for iOS or Android made it feel as good a time as any to walk away and take my chances on building something that I actually want to use in the technology that I like to use. Will I ever be back at Pivotal Labs? Probably, though I'll be making sure I choose an office on the frontier, working with companies in a new market, as far away from a CloudFoundry-focused office as I can get.
EDIT: A note about management in Labs: The communication within Labs itself has always been great; very little bullshit, very little waiting to find out something that is important to you, if someone can help you find it out. The managers for the devs at Labs are all either current or former Labs developers (some are now full-time management, some are devs for CF, and the majority still consult 36 hours a week). They understand their developers very well, and to a person would not let such an incident happen under their watch. That said, CloudFoundry and the Pivotal top level organization have entirely unrelated (and unknown to me) management structures, the bulk of which was likely inherited from EMC (as were the PR/Marketing/HR/ and some other departments.)
- GBond 12y agoThanks for a thoughtful comment that provided way more information and classiness then the OP's