6th > December > 1998 Archive

Analysis: How Bill Gates discovered the Web

What did Microsoft think about the Web, and when did it think it? Establishing the truth of this is of vital importance to the DoJ's case and to Microsoft's defence. Bill Gates says the company started work on browsers in April 1994, just before Netscape was incorporated, but Microsoft documentation from 1995 shows the company still formulating its position, and Gates himself didn't go public with the Microsoft 'embrace the Web' strategy until late that year. Gates claimed in August of this year that he'd pinpointed the date as 5 April 1994, when he'd said at an executive retreat: "Hey, we're going to get it integrated into the operating system," and that on 16 April he'd given executives responsibility for developing browser technology for Windows. But by May 1995 it appears that plans weren't going particularly well; DoJ trial exhibits 20 and 21 show on the one hand Gates delivering a 'road ahead' directional statement to his executives, and on the other Ben Slivka detailing the chaotic state of Microsoft development, and proposing a strategy for a Microsoft-owned "SuperWeb." The Gates document (exhibit 20, dated 26th May 1995) is entitled The Internet Tidal Wave, and provides a picture of what Gates was thinking about Microsoft's strengths and weaknesses at the time were, how the company could move forward, and who the big competitors and threats were. He notes that competitors' Web sites are better than Microsoft's and that "Sun, Netscape and Lotus do some things very well. Amazingly, it is easier to find information on the Web than it is to find information on the Microsoft corporate network". At the time one of Microsoft's strategies for the Web was to incorporate browsing capabilities into its applications, and although Slivka (we'll get to exhibit 21 another day) was simultaneously pointing to this as being a blind alley, Gates doesn't yet seem ready to drop it. "All work we do here can be leveraged into the HTTP/Web world. The strength of the Office and Windows businesses today gives us a chance to superset the Web. One critical issue is runtime/browser size and performance. Only when our Office-Windows solution has comparable performance to the Web will our extensions be worthwhile. I view this as the most important element of Office 96 and the next major release of Windows." Office 96 became Office 97 when it shipped, while the next release was Windows 98. Microsoft had to some extent been trying to view the Web as territory it could extend its file formats into -- hence the association in "Office-Windows solutions". Size and bandwidth problems were clearly starting to show this wouldn't work, and although much of the world had already embraced Microsoft file formats for productivity applications, Microsoft was only just beginning to twig that the Web hadn't done so, nor was it about to. But Gates is getting there. Although at the time Microsoft was on the point of launching an online system, the Microsoft Network (MSN), he suggests that approach may already have been superseded. "The online business and the Internet have merged. What I mean by this is that every online service has to be simply a place on the Internet with extra value added. MSN is not competing with the Internet [he's almost there, but not quite], although we will have to explain to content publishers and users why they should use MSN instead of just setting up their own Web server." This is exactly the problem that faced the established online systems like AOL and CompuServe at the same time, and they weren't necessarily coming up with the right answers straight away. But as we see from Gates, they were in good company -- in May 95 Microsoft knows it has to deal with the Web, but thinks it can still go with a proprietary online system as well. So why use MSN? "We don't have a clear enough answer to this question today. For users who connect to the Internet some other way than paying us for the connection we will have to make MSN very, very inexpensive -- perhaps free." Note the significance of "inexpensive" -- one of the reasons Microsoft didn't want to let go of MSN entirely at that point was because it felt it could associate billing (if you'll pardon the expression) systems with it. The other online systems too were at the time trying to wrap their heads around how they billed or microbilled instead of giving stuff away for free. Now, note the recently unveiled Microsoft 'TV' model used as a justification for giving IE away. TV stations broadcast programmes for free, and recoup the cost in advertising, and that's what Microsoft says it started doing when it started shipping IE for free later in 1995. But here's what Bill thinks in June: "The amount of free information available today on the Internet is quite amazing. Although there is room to use brand names and quality to differentiate from free content, this will not be easy and it puts a lot of pressure to figure out how to get advertiser funding." Further on in the document he makes it clear he's not letting go of the MSN approach yet. "We need to determine a set of services that MSN leads in -- money transfer, directory, and search engines. Our high-end server offerings may require a specific relationship with MSN." We think this means linkage between NT Web server sales and the purchase of MSN services, but no doubt Bill can't remember what he meant, these days. The Competition Gates identifies some familiar, and some not so familiar names here. "Our traditional competitors are just getting involved with the Internet. Novell is surprisingly absent... however... Novell has recognised that a key missing element of the Internet is a good directory service. They are working with AT&T and other phone companies to use the NetWare Directory Service to fill this role." The AT&T project for a parallel and proprietary Internet was later abandoned, although today Novell is trying hard to leverage NDS on the Internet. "All Unix vendors are benefiting from the Internet since the default server is still a Unix box and not Windows NT, particularly for high-end demands. Sun has exploited this quite effectively." Other competitors are obvious because of the file formats, and it seems Bill is starting to get it here: "Browsing the Web, you find almost no Microsoft file formats. After ten hours browsing, I had not seen a single Word DOC, AVI file, Windows EXE (other than content viewers), or other Microsoft file format. I did see a great number of QuickTime files... Another popular file format on the Internet is PDF, the short name for Adobe Acrobat files... Acrobat and QuickTime are popular on the network because they are cross-platform and the readers are free." Then there's Netscape. "Their browser is dominant, with 70 per cent usage share, allowing them to determine which network extensions will catch on." Gates is worrying about the standards escaping from Microsoft control. "They are pursuing a multi-platform strategy where they move the key API into the client to commoditise the underlying operating system. They have attracted a number of public network operators to use their platform to offer information and directory services. We have to match and beat their offerings including working with MCI, newspapers, and others who are considering their products." Here Gates is still thinking in terms of online systems and 'pay-per-view' models. But can we see a 'get Netscape' strategy emerging? If the control moves out from the client onto the Web, with maybe Sun servers dominating the back end, what happens then? "One scary possibility being discussed by Internet fans is whether they should get together and create something far less expensive than a PC which is powerful enough for Web browsing. This new platform would optimise for the data types on the Web. Gordon Bell and others approached Intel on this and decided Intel didn't care about a low-cost device so they started suggesting that General Magic or another operating system with a non-Intel chip is the best solution." Ah yes, so before Oracle invented the NC and Intel (originally) turned it down, the idea was floated past Intel, and Gates knew about it. Around about this time Intel was engaged in its messy battle with Microsoft over NSP, so maybe in a parallel universe Intel jumped the other way, and is three years ahead of where our Intel is in appliance design, and operating software. Slivka is also worried about this -- he thinks Siemens and Matsushita might turn out to be the troublemakers. What Bill wanted done He faces two ways here, to some extent. Slivka identifies lack of co-ordination of efforts at Microsoft as a major problem, but you can see how this arises. Bill says "I want every product plan to try and go overboard on Internet features", and this basically drives Microsoft's different product groups further along the road of multiple overlapping efforts. But: "One element that will be critical is co-ordinating our various activities... Paul Maritz will lead the Platform group to define an integrated strategy that makes it clear that Windows machines are the best choice for the Internet. This will protect and grow our Windows asset. Nathan [Mhyrvold] and Pete will lead the Applications and Content group to figure out how to make money providing applications and content for the Internet. This will protect our Office asset and grow our Office, Consumer and MSN businesses." Note we're facing two ways here as well. Maritz seems to be elected to 'invent' the integration strategy, or at least to build on what unco-ordinated shreds of it already exist, while Mhyrvold, old Advanced Technology Group trusty and the man who drove the early Microsoft apps strategy (which was incidentally tanked by Lotus in the early 80s), is still trying to link Office, MSN and cash together into a winning formula. This approach was actually what Microsoft was talking publicly about at the time (eg. Steve Ballmer in an InfoWorld interview of the time: "I tend to think of [the Microsoft Network] as part of the extra-enterprise pitch that we make to every corporate account that we call on.") Gates, meanwhile, has several things he wants done on the Windows platform. "We need to understand how to make NT boxes the highest performance HTTP servers. Perhaps we should have a project with Compaq or someone else to focus on this... We need a clear story on whether a high volume Web site can use NT or not because Sun is viewed as the primary choice." Three years on, this "clear story" has yet to ship in its entirety. "We need to establish distributed OLE as the protocol for Internet programming [oops...]. A major opportunity/challenge is directory. If the features required for Internet directory are not in Cairo or easily addable without a major release we will miss the window to become the world standard in directory with serious consequences [oops again]. Lotus, Novell and AT&T will be working together [not, as it transpired, for long] to try and establish the Internet directory. Actually getting the content for our directory and popularising it could be done in the MSN group." Clearly, Bill still hasn't got it. He's still thinking in terms of getting Microsoft standards out there and pulling mindshare and dollars away from the existing Internet, and he's thinking of similar proprietary projects which never flew as the primary competition. But the client strategy is emerging. "First, we need to offer a decent client (O'Hare)... However, this alone won't get people to switch away from Netscape. We need to figure out how to integrate Blackbird, and help browsing into our Internet client." So we have a possible smoking pistol here. O'Hare (how you get out of Chicago, aka Windows 95) was, as Microsoft claimed later in the DoJ actions, intended to be integrated. But it didn't exactly include browsing (as shipped, it included MSN browsing), and Gates is first talking about doing that and about switching people away from Netscape in May 1995. The proprietary Blackbird strategy was failing at the time, and the Internet hadn't moved centre-stage in Microsoft's product plans. "We need to move all of our Internet value added from the Plus pack into Windows 95 itself as soon as we possibly can with a major goal to get OEMs shipping our browser preinstalled. This follows directly from the plan to integrate the MSN and Internet clients." So on the other hand the integration strategy may be starting to take form, but it's doing so in the same breath as the N word. Not that Microsoft doesn't have a browser strategy -- it has several, possibly helped by Gates being a little vague about what a browser is or could be: "We need to determine how many browsers we promote. Today we have O'Hare, Blackbird, SPAM MediaView, Word, PowerPoint, Symettry, Help [this one indistinct on the DoJ printout, but we think he means Help] and many others. Without unification we will lose to Netscape/HotJava." And again, integration starts to take shape: "Over time the shell and the browser will converge and support hierarchical/list/query as well as document with links viewing... We need to establish OLE protocols as the way rich documents are shared on the Internet. I am sure the OpenDoc consortium will try and block this." Quite... "We need to give away client code that encourages Windows-specific protocols to be used across the Internet" -- Bill clearly knows what he wants to do, even if he's initially choosing the wrong ways to do it. If Windows-specific protocols become Internet standards, then Windows is the client, and NT the server for the Internet. And he's also figured out the possibilities of thin client/server two years ahead of trying to do anything about it. "We should also consider whether to do something with the Citrix code that allows you to become a Windows NT user across the network." And more on file formats: "We need to make sure we output information from all of our products in both vanilla HTML form and the extended forms that we promote... We need to decide how we are going to compete with Acrobat and QuickTime since right now we aren't challenging them. It may be worth investing in optimising our file formats for these scenarios." And the future? "I believe the work that has been done in Consumer, Cairo, Advanced Technology, MSN and Research position us very well to lead... The electronic world requires all of the directory, security, linguistic and other technologies we have worked on... There will be a lot of uncertainty as we first embrace the Internet and then extend it... We have an opportunity to do a lot more with our resources. Information will be disseminated efficiently between us and our customers with less chance that the press miscommunicates our plans [aw, shucks...]. Customers will come to our 'home page' in unbelievable numbers and find out everything we want them to know." Of course, if three years later masses of your internal documentation turns up on a DoJ Web site, this doesn't necessarily apply. ® Complete Register trial coverage
John Lettice, 06 Dec 1998

Sun set to bring Linux to UltraSparc

Sun is set this week to embrace Linux as an operating system for low-end servers, if reports in the US edition of Computer Reseller News are to be believed. According to a report in this week's edition of the paper, Sun has spent most of the year porting Linux over to its UltraSparc CPU, and will this week announce widespread support for the free OS on its hardware. An odd move, you might think, given Sun already has a Unix knock-off of its own, Solaris. But therein lies the rub. As Linux continues to make headlines in the more mainstream IT press, and it garners more interest from the wider IT community -- we won't say supporters until there are some hard userbase stats to back the assertion up -- it is proving more of a threat to Unix operators like Sun. True, Linux is traditionally perceived as a threat to Windows NT, and that will continue to be the case, but it doesn't rule out the danger to the commercial Unix market, particularly at the low end, that the free OS also poses. And no one more clearly controls the low, volume end of the commercial Unix market than Sun, which has done very nicely, thank you out of flogging cheapish servers to smallish business, such as small ISPs, Web-hosting/design outfits and repro bureaux, who need a tough server environment to which they can connect a handful of clients of various platforms. Companies like these, however, are looking at Linux with keen interest. It may not be up to handling high-bandwidth server applications like multi-site hosting and centralised databases, but its more than capable of running DNS servers, firewalls and low-traffic Web site hosts on cheap Intel and Mac hardware. That worries Sun, for all its talk in the CRN article of considering Linux users "in the open standards camp" and therefore "making good things happen", and bullish talk from Robert Novak, Sun's power workstations group manager, of the free OS offering "a real strong alternative to NT and Windows". Hang on a mo' -- wasn't Solaris supposed to do that? Actually, it is, and we suspect Sun doesn't believe a word of it either. The Linux-on-Sun project's codename says it all: UltraPenguin. Sure, it combines the name of Sun's processor with Linux's Antarctic mascot, but it's hardly inspiring is it? Sun is, after all, the company of Java, of Jini -- in short, the company of the groovy codename that makes the bigtime. No, it's a banal codename for a project that's more about business than open systems evangelism. Linux on UltraSparc is there to shore up Sun's low-end against Linux on Intel. And, for that matter, NT on Intel, since the Microsoft OS too lies largely below the massive multi-processor applications that Unix is better suited to. And then there's the Netscape connection. Sun will, of course, soon be taking over the browser Satan's back-end software business in exchange for its support for the AOL takeover. Netscape dipped its toe in open source waters earlier this year, and its recent backing of Linux distributor Red Hat suggests it didn't find that water too cold. Maybe it now has a large selection of its server-based software running on Linux. Sun is obviously going to have to do something with it. Had the UltraPenguin project never been initiated, Sun would probably have killed Netscape's Linux apps. It may still do so, but it makes sense, if you're going to promote Linux on UltraSparc, to have some solid application software other than Apache to back it up. One quick port later and you have a software suite you can bundle with your own Linux distribution but force Intel Linux users to pay for (don't forget we're talking here about mainstream users who are used to paying for application software, provided it works out of the box). ®
Tony Smith, 06 Dec 1998

Eurospook plan for Web and wireless bugs

A global wire tapping system covering mobile phone, fixed wire, Internet and satellite is being pushed by the European police co-ordination body, Europol, according to German Web magazine Telepolis. A paper published by the mag (much of the meat in German, sorry people) purports to be a high level Europol document putting forward proposed legislation for Europe, and trailing the prospects of links with US and other non-European security agencies. But if the document, entitled Enfopol 98, is genuine, Europol has moved way beyond its brief. Conspiracy theorists might say that's what you'd expect of a transnational bunch of spooks, but it far exceeds existing European policy on transnational policing and electronic surveillance. "Exactly," as the conspiracy theorists would have it. On the other hand, Enfopol 98 takes numerous European and international initiatives to deal with drugs, terrorism and cybercrime to their logical conclusion. It is claimed to have been drawn up in the form of a "Draft for a Recommended Resolution" by Europol at the EU's behest (Telepolis doesn't specify who within the EU) in order to "simplify the passage of the resolution." The draft does not seem to have appeared at Brussels yet, but that's not to say it won't. Enfopol 98 plans to have tapping built into all communications systems. Each country will have "interception interfaces" in telephone exchanges, ISPs, cellular networks and satellite ground stations. Intriguingly Telepolis alleges that the Iridium ground station in Italy is slated as one of these. It's not known what Iridium's views on this are, but we've heard some interesting suggestions about the usefulness of global commercial satellite phone systems to US security services. Tapping data will be shared across countries - each state should be able to tap into communications right across the EU, and the system is intended to extend to the US, Canada and Australia. The bona fides of the document are also reinforced by the suggestion that an identification system for individuals should be tacked onto it. Data to be exchanged between countries would include names, addresses, phone numbers, credit card numbers, email addresses and computer passwords. Only a crazed cop would think these could be obtained or could be rammed through in legislation. A report by Duncan Campbell in today's Observer newspaper takes the paranoia to fever pitch. Duncan may, like a substantial proportion of The Register, hail from god's own country, but he shows frequent signs of being barking. According to Campbell Enfopol 98 will be brought in as part of the European Convention on Mutual Legal Assistance, and will go into national law by 2000. The sinister bits, says Campbell, will be hidden in a "technical handbook" which is (was) unlikely to be read by ministers. You can see why Europol would want such a system, and why the UK's own dear home secretary Jack Straw (currently trying to figure out a plausible strategy for springing a Chilean mass-murderer) would love to be able to help. You can also see why the bug-happy French security services, who want to read everything, and the control-freak German and Italian ones, who've a long history of urban terrorism, would support it too. But how do you get it through? And how does it work? A system that knows where everybody is, what they're saying, and how to get into all of their email is a global civil liberties no-no. Even if you got it through, all hell would break loose when the people found out (this, incidentally, is why Jack Straw hasn't yet dared to send Pinochet to Northolt airbase, where a Chilean military plane awaits him). Enfopol 98 also requires real-time bugging capabilities. If, say, the Italian security services were bugging a mobile phone in the UK they'd want to do it real time, and just dealing with non-encrypted systems would be tricky, and encrypted GSM very tricky indeed. The networks will squeal if they have to pay for it. ISPs? It's possible to bug by forcing all the major ISPs to implement interception systems, but unless you legislate to licence ISPs, you can't get them all. Then you have to legislate to restrict the available of leased lines (license these as well), then you start to worry about how you intercept calls dialled direct into computers. And then you've got to worry about how you actually read the traffic. Enfopol says codes have to be broken, but it's not clear how. Government agencies don't have any obvious abilities to do this, and although the major software companies would no doubt roll over (not a lot of choice), they don't necessarily cover the more interesting parts of the waterfront. So even if the whole shebang comes in globally, you can still operate secure communications point to point or via the Web (unless of course they then try to make it illegal to transport messages they can't read). It can only work if they can monitor everything, all the time, in real time. And they can't. Europol itself, incidentally, was set up to deal with drugs, and is probably the "high level Group, set up following the Dublin Council, [which] is finalising an Action Plan to fight cybercrime." (EU documentation) Europol's responsibilites have now been extend to include: illicit trafficking in nuclear or radioactive substances; crime linked to clandestine immigration networks; trafficking in stolen vehicles; and trafficking in human beings (sexual exploitation). You can see why it might think it needs the extra surveillance, can't you? ®
John Lettice, 06 Dec 1998
Dixons Carphone Warehouse

Dixons threatens legal action against Intel, Fujitsu

Dixons Stores Group will sue Intel and Fujitsu unless they withdraw allegations that the retailer is overcharging PCs, according to a UK Sunday newspaper.
Drew Cullen, 06 Dec 1998