Wot? Me (Nick Shelness) Biased?

I recently received an email from an independent reader who accused me of manifesting a pro-Microsoft, and an anti-IBM, bias. It set me to thinking. I've never worked for Microsoft. I spent part of my career working for one of its major competitors (Lotus). I had the honor of being an IBM Fellow. There are many things that Microsoft has done over the years of which I do not approve. So que pasa?

I guess the key point is, that I have never taken the approach that because something comes from Microsoft, it must be good. Similarly, I have not taken the approach that because something comes from Microsoft, it must be bad. Ditto with IBM. I call things as I see them, and if I have a bias, it is a technical bias. I have a warm spot in my heart for elegant architectures, and a disdain for vacuous marketectures.

Cutting to the chase ...

Thoughts on Microsoft

  1. I am incredibly impressed with Microsoft's multilevel approach to cloud computing, and the largely seamless way that this approach bridges the gap between on-premise and cloud-based services. There are elements of this approach that are self-serving--Microsoft has a strong, and strengthening, position in on-premise Windows Server-based services. One would expect Microsoft to try and extend this position to the cloud. That it has chosen to do so in an elegant fashion garners my respect. It is interesting that it is an ex-Loti--Ray Ozzie, one of the world's uber architects and coders--who appears to have spearheaded this effort, and that Dave Cutler--Microsoft's (and DEC's before that) ultimate uber coder--wrote the Azure kernel.
  2. I am incredibly impressed with Microsoft's work on Identity and its recent refashioning of a badly broken approach to open interfaces. Yet again, this effort has not been some corporate initiative; it was pioneered by a deep thinker--in this case Kim Cameron.
  3. I am incredibly unimpressed with Microsoft's historic approach to "standards." Embrace, extend, patent, and license was not a useful approach for either Microsoft or its customers. Siting responsibility for Microsoft's standards activities in the Microsoft legal department was a bad idea. Microsoft chose not to participate in the Open Document Foundation. This was dumb, and allowed its competitors to promulgate a standard (ODF) that left Microsoft Office at a disadvantage. I was once told that the first law of politics is "Be There." Microsoft was not. This was followed by a cack-handed attempt to ram Microsoft's ODF alternative through the ISO standardization process. Microsoft now appears to have, at least partly, recognized the error of its ways, and to have joined the ODF. This may be too little, and too late, but it is certainly a move in the right direction.
  4. I have been a relatively happy Windows Vista user. I run it on two hulking (over-clocked Core2 Quads, 8GB of over-clocked RAM, 2-x-RAID0 Solid State Disks (SSDs), 2TB of RAID5 storage, 64-bit desktops, and on two 32-bit laptops--one a Vista-ready Core Duo with a 64GB SSD (32GB was too small) and a 200GB HD, and the other an undersized old ThinkPad. That being said, they are flaky. It's not so much a case of Blue Screens of Death (BSODs) or hangs, as much as performance grinding to a halt for no apparent reason, or a system failing to restore from hybrid sleep. My sense is that Vista was too much about trying to errect barriers against Linux and Samba than meeting customer needs. My sense is also that Windows 7 is a much more customer-focused release. I plan to try it in beta soon.
  5. I'm not a great fan of the Ribbon in Office 2007. When it comes to UI controls, I remain a fan of the Lotus Info-Box. I found it easy to navigate its tabs, and any changes took immediate effect. I still use a bunch of old 123 spreadsheets occasionally, and navigation is never a problem. Though I use Word 2007 at least once a week, I find myself having to resort to Help far too often when trying to find some feature.

Thoughts on IBM and Specifically Lotus

  1. IBM did some really smart things soon after acquiring Lotus. John Thompson, then the head of IBM's Software Group, accepted Mike Zisman's (Lotus CEO) and Jeff Papows' (Lotus COO) argument that Notes had to be sold aggressively as a Messaging+ solution, and that this would require a significant influx of cash. Prior to this, Lotus had sold barely 1,000,000 Notes licenses. It was this commitment on IBM's part that allowed Notes to run Microsoft such a close race, especially in the enterprise space. It is worth noting that the pairing of Mike and Jeff was one of those cases in which 1 + 1 = more than 2. It is a case of much regret that Mike's decision to put his family (and his ex-wife's cancer) first, and Jeff's enforced departure, left Lotus with a leadership hole from which, in this author's opinion, it has never fully recovered.
  2. Since the departure of Mike and then Jeff, Lotus has seen a succession of general managers (Al Zollar, Ambuj Goyal, Mike Rhodin, and now Bob Picciano) on two-year rotations. Each of them seems to have gone in for a change of strategy at the end of their first year, which lasted until it changed again at the end of their successor's first year. I have no objection to an evolution in strategy as long as it is soundly based and technically achievable. The trouble with the vast majority of IBM's general managers is that they are "jacks of all trades and masters of none." The small minority, who manage to be "masters of many"--like Sam Palmisano, IBM's current CEO, and John Kelly, IBM's SVP of Research--get to join the management committee and run the company. The other problem is that many of them "don't know what they don't know" and aren't interested in finding out. In some cases, this doesn't matter. Senior architects, who are IBM lifers, and who don't rotate (Don Haderle at DB2 comes to mind), effectively set technical strategy, as general managers come and go. At Lotus there seems to have been no suitably wise technical leaders to question some of management's more harebrained strategies, or if they did, they were ignored.
  3. When Lotus does smart things, I will applaud the company. The move to Eclipse and the adoption of Xpages are two good examples. But they need to be followed through. Additional support for native (composite-application) Eclipse-based Notes applications was striking in its absence at Lotusphere 2009. When Lotus does dumb things, like Workplace, attempting to slide DB2 under Domino; and now announcing LotusLive with Outblaze, as opposed to Domino-based, email, I will call it dumb.

If you disagree with any of the above, or have an additional comment, please post it.

... Nick Shelness

Post a comment

You must be logged in to post a comment. To comment, first join our community.