From owner-freebsd-hackers Sun Dec 10 09:21:16 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA25094 for hackers-outgoing; Sun, 10 Dec 1995 09:21:16 -0800 (PST) Received: from etinc.com (et-gw.etinc.com [165.254.13.209]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA25089 for ; Sun, 10 Dec 1995 09:21:13 -0800 (PST) Received: from et.htp.com (et.htp.com [199.171.4.228]) by etinc.com (8.6.11/8.6.9) with SMTP id MAA28497; Sun, 10 Dec 1995 12:35:40 -0500 Date: Sun, 10 Dec 1995 12:35:40 -0500 Message-Id: <199512101735.MAA28497@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Version 2.0.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: "Jordan K. Hubbard" From: dennis@etinc.com (dennis) Subject: Re: FBSD support inc. Cc: julian@freefall.freebsd.org, hackers@freebsd.org Sender: owner-hackers@freebsd.org Precedence: bulk Jordan writes.... >Julian Elischer writes... >> I've been thinking about setting up a company (actually more than one) >> to give commercial support for freeBSD. > >I think this is a fine idea, and not exactly one we haven't been over >before, but I still wouldn't want to start with quite so complicated a >picture as this. > >When you're selling support, you're selling a service that *has* to >work for each and every customer who's forked over good money for it >or the whole thing breaks down and you've got the better business >bureau calling you on the phone (or worse, somebody's legal dept). >This means that it's got to be possible to quantify just how long any >given service call will take to respond to and also to make sure that >each and every call is tracked and resolved properly. Even simply >measuring this is non-trivial, and the difficulty increases >exponentially with the number of staff (and cubed by their distance >from you). You also want to be able to easily determine where you're >spending the most money and time since support will eat you up (and >make you unprofitable) very quickly if you don't keep things carefully >streamlined. I want to point out some of the less obvious issues with such an endeavor, not just in providing the service but regarding the impact that it might have on the image of FreeBSD. There are clear problems with the whole concept, and I also think that you might by overestimating the value of support for a non commercial product. This issue with "free" products is not necessarily that there is no support, but that there is no reasonable expectation that all included features of the product will work well, and that there is a questionable future and no business plan for the product. Basically, the issue is that there are no guarantees that anything will work, and I don't think that your "support" proposal will change that in any way unless you are willing to support (and be willing to take the heat) when virtually anything doesn't work. A real danger is that the "free" list support that is now available (and quite good) will diminish because many of the good contributors will be bound to a commercial obligation. This could hurt the positive word of mouth that points out that the list support for FreeBSD is in many ways better than the commercial support provided by other UNIX vendors. There is also a widespread understanding that UN*X support generally sucks, even when you pay a LOT for it, and its certainly no worse than what you get from Microsoft with their new 900 numbers. Success is much more about marketing than it is support. (i.e. Microsoft) The fact that support exists is a good marketing issue, but I don't think that people who are trying FreeBSD are junking it because the support is no good. Substantial hand-holding is a thing of the past....if the product doesn't work out of the box you basically lose. Remember that 3rd party contracts are only generally purchased AFTER the customer decides they will use the product. In order to be successful, you will have to have your own release (Caldera, Red Hat...sound familiar?) or count on other people to get the customers up and running....which is by far the most time-consuming aspect of support. Remember that virutally all commercial products include installation support....how will you handle this for a "free" product? I think that a product is either commercial or its not commercial, and unless you can guarantee that some set of features works and will continue to work, you will have difficulty having much impact. The bridge from non-commercial to commercial involves a lot more than just getting a bunch of people together with a dial-up or ISDN connection. Remember also that you will be supporting much more than just a SCSI driver or a mail reader, you will be expected to support complex networking issues that require much more than a single machine in someone's basement. People will want to know why BGP4 on gateD doesn't work well with a wellfleet router, and why their ISDN connections drop off intermittently when and only when connecting to some obscure version of NT server. They won't be happy if you tell them that one or more of their issues isn't supported, even if its well mapped out in a contract. We have 30 PCs in our lab, Pentiums, 486's, 386's with LAN monitors, serial data scopes, various CSU/DSUs and modems, 10 baseT, BNC, AUI, every kind of cable, etc, etc just to support a couple of WAN cards (and Jordan wants to know why they cost more than $120.). Support is no simple task. It is by far the most time consuming aspect of our business. Designing cards, software....easy stuff. But once you make a sale you're obligated to help that customer no matter how stupid and annoying he or she is. Just be prepared. Dennis ---------------------------------------------------------------------------- Emerging Technologies, Inc. http://www.etinc.com Synchronous Communications Cards and Routers For Discriminating Tastes. 56k to T1 and beyond. Frame Relay, PPP, HDLC, and X.25