spf-discuss
[Top] [All Lists]

Re: Changing Standards - Flashback

2004-10-06 23:06:12
Hector,

At 08:50 PM 10/6/2004, you wrote:
Back in 1992/93, I went to the Microsoft Software Developer Conference at
Seattle.  I arrived at the hotel around 5-6pm. I went to my room, exhausted
from long Miami to Seattle trip,  I turned on the TV and laid down.  I got
excited when a local Seattle news station came on with an interview with
Bill Gates about the upcoming/next day MS Software Developer Conference.

I recall a statement that stayed with me till this day when he was asked how
he deals with competitors and how does MS stay on top as the market leader.
His answer was, honest to god,  no lie, this is what he said:

        "We stay ahead of our competitors by changing the standards every
two years."

Might it be argued that Mr. Gates' purported statement above was not referring to standards in the sense we discuss here on this list, but more as a marketing statement? In other words, to a marketer, raise the bar, up the ante, change the standards might all mean roughly the same thing. Arguably, "up the standard" rather than "changing the standards" might have been a better way to phrase, but when one is preparing for a big event at late hours, the mind gets a bit exhausted.

When you think about it, Microsoft did and still does try to get out new releases of software every two years or so. If memory serves, that was especially so during the early 1990s time frame with their application suite and OS releases.

I hope nobody holds me to words I might utter to media or anyone else when near exhaustion just before a major presentation for key partners for projects my company is working to roll out. Especially so when those utterances might be dredged up from a time during a ramping up period well over ten years in the past.

This is not bashing. It is just a scary thought about MS that has always
stayed with me.  Its on tape in the local tv news station achives if proof
is required.

Nobody is perfect and not everything you see or hear on tapes from long ago ring fond memories to their creators. Come to think of it, I have some tapes from a local Denver business talk radio show I produced and hosted for two years back in the late 1990s and possibly some master tapes of interactive satellite video broadcast training I developed the model for and hosted in a past life that I should probably go find and burn now... ;-)

How this related to SPF?  Well,  I hate to be in a position where SPF will
be changing every year or every 2 years!  Who will own the specs and
standard definition?  Meng?  Under what entity?  Or will it be Microsoft
when SPF2 begins to promote SenderID?

Why would such things take place? The Internet participants themselves won't permit it. Changes are generally adopted only if sufficient value added is perceived by adopters in making them. Once I publish the basic SPF DNS TXT records that suit my company's needs for domain identity protection in email, only a really big change (e.g., wide spread support for SPF RR records) is probably going to make me go back and have records adjusted in company DNS zone files.

I think that the challenges ahead for SPF will primarily stem from being able to respond to more of the corner case needs. In some situations, as unfortunate as it may seem, the best answer for some may be, "if you wish to exchange email in the future with others, you can't do what you have been doing anymore because it compromises security or other needs upon which nearly everyone else on the Internet depend".

Large companies are not inherently evil and they all started out as small companies and/or small group operations within larger companies. The biggest asset large companies supporting an open standard bring to the table is their reach and consequential ability to pervasively promote such a standard. Most of these companies and others who participate on the Internet understand that standards burdened by ownership rights and licences are probably not really standards at all.

The expression "armies of lawyers" coming to mind here may be fitting. The job of armies is to go forth, break things and kill people. Corporate counsel around here has said that business people make deals and lawyers break them. I suppose that concept is somewhat self evident, but it does underscore why open standards seem to make sense in an overly litigious world. For the record, our corporate counsel may have indeed gone forth and broken a few deals (done only in the best interests of the company), but has not actually killed any people to my knowledge or to the best of my recollection. ;-)

As regards proprietary property, by way of example, perhaps you might remember what happened when Microchannel architecture was introduced by IBM back in the early days of the personal computer. The rest of the industry answered that proprietary licensed architectural growth path solution with EISA, an open architectural standard. Microchannel, though a good architecture, failed to achieve its goals for IBM and the world moved on. Large companies like IBM often do learn from mistakes and I think, in part, that is why IBM has become such a proponent of embracing open standards today and also think that they are doing well for themselves and others in making such moves.

On the Internet, it seems that the most productive and useful applications are best accomplished by common agreement, what some call open standards of interoperability. I think this always has been true and probably always will.

History records the contributions people and companies make to open standards on the Internet and that really should become its own reward for the participants. It is what people and companies build atop these standards that allows them to profit or make further contributions from as each sees fit.

Sincerely,

Hector Santos, CTO
Santronics Software, Inc.
http://www.santronics.com
305-431-2846 Cell
305-248-3204 Office

Best,

Alan Maitland
The Commerce Company - Making Commerce Simple(sm)
http://WWW.Commerco.Com/




<Prev in Thread] Current Thread [Next in Thread>