There is great interest in Microsoft and its policy with Internet Explorer, and some interesting comments from Michael Toy of Netscape about "We are working hard to figure out not only how we make the source available, but to also really get out of the way of a net-community based effort to make great client-side software". This has been interpreted by some as saying that Netscape will release the source code for V 5 of their browser, and then hand over further development to the community. Anticipating that the chemical community might in the medium term have to rely on Internet Explorer only, I decided to see how Version 1 of Chime would work with the latest released version of IE 4.0. Microsoft, whilst they had <10% of the browser "market", took at least some care to be "compatible" with Netscape, and this included supporting the Netscape plug-ins. The Chime plug-in mostly worked with IE 2 and 3 (although it was flaky, as the developers of Chime themselves pointed out). All you had to do was to insert Chime into the IE plug-ins folder, and it (sort of) worked. Now that Microsoft reputedly have >50% of the browser market, they appear to have a rather different attitude. It may be a salutory lesson for the future. Firstly, let me say how Netscape handled plug-ins. By installing a plug-in, the user automatically had configured for them all the appropriaate MIME types supported by that plug-in. This included lets say chemical/x-pdb for a typical chemical MIME type. Microsoft handle MIME in a much more complicated way. On Windows, its via the system wide registry. On a Mac, they attempt to achieve such a system wide configuration by using a separate program called Internet Config. What follows here is specifically for the Mac. I have not yet done this on a PC. When I installed IE 4 on my Mac, I specifically asked that all the preferences be imported from Netscape rather than using Internet-config 1.4 (which does not recognise plug-ins). This should have included configuring IE 4 to use the Chime plug-in. I had carefully first installed a copy of Chime in the IE plug-in folder before starting it. So what in the end happened? IE ignored all the Netscape chemical MIME types, and despite being asked not to, defined the (two only of about 10) chemical types defined by default by Internet Config. These were not associated with the Chime plug-in but instead the option set was to "save to disk". Now, IE DOES offer a reconfiguration option to instead resolve the file to a plug-in. If this is set manually to Chime, and IE 4.0 is offered a pdb file, it again refuses to use Chime, and instead IE cycles in a hung loop. Arguably, all of this might just be an "oversight" or "bug" in IE 4.0, or a newly introduced incompatibility with Chime 1.0 which might be fixed at version 2.0. But it does mean that a chemical use with a Mac trying to migrate to IE might be tempted to give up with Chime, and find it easier to say %$! to plug-ins, they are only ropey old Netscape technology anyway. Presumably, if we REALLY care about chemistry, we will re-invent the web the Microsoft way using ActiveX (at our own time and expense) and all will work again. On a PC. Probably not on Unix or Mac. Or maybe the future is with XML custom solutions, of which CML (chemistry markup language with Peter Murray-Rust's JUMBO CML viewer) is one example. Microsoft, I might add, are ardent adoptees of XML. Perhaps the common denominator that the chemical community was briefly able to achieve during the period 1996-1997 with the "old style" web browser and a plug-in such as Chime was just a temporary phenomenon. Arguably, one might look back to that period as a golden, simple one where many people could be molecularly enfranchised with little effort. We may never see its like again. Henry Rzepa. +44 171 594 5774 (Office) +44 594 5804 (Fax) +44 575 1839 (Home) http://www.ch.ic.ac.uk/rzepa/ chemweb: A list for Chemical Applications of the Internet. To post to list: mailto:chemweb@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/chemweb/ To (un)subscribe, mailto:majordomo@ic.ac.uk the following message; (un)subscribe chemweb List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)
On Sun, 25 Jan 1998, Rzepa, Henry wrote:
There is great interest in Microsoft and its policy with Internet Explorer, and some interesting comments from Michael Toy of Netscape about "We are working hard to figure out not only how we make the source available, but to also really get out of the way of a net-community based effort to make great client-side software". This has been interpreted by some as saying that Netscape will release the source code for V 5 of their browser, and then hand over further development to the community.
The press release is at: http://www.netscape.com/newsref/pr/newsrelease558.html?cp=nws01flh1
Or maybe the future is with XML custom solutions, of which CML (chemistry markup language with Peter Murray-Rust's JUMBO CML viewer) is one example. Microsoft, I might add, are ardent adoptees of XML. Perhaps the common denominator that
I am presenting a (free) virtual lecture on this next week sponsored by Chemweb Ltd. (no connection) and VEI ; 1998-02-04:16:00 UTC. [Many of you may have attended Henry's virtual lecture - the first in this series - last month.] In this I shall present the basis of XML and its advantages for producing WWW solutions. A new snapshot of JUMBO will be released in alpha this week, and a beta to coincide with the virtual lecture. [JUMBO will *not* be used for the lecture itself, although I see this technology being ready in about 3-6 months time. A major factor will be to what extent browsers support reasonable performance in a platform-independent manner, epsecially Java.] I cannot comment on Netscape's future pplans for XML, but they have been active members of the XML community. Assuming V 5.0 is GPL, then I would expect that high quality XML tools will follow very rapidly (there are already freeware parsers, and browsers are following.) The key difference between plugins and XML (in this area) is that XML supports structured documents. In this way it is possible to integrate "documents" and "data" in a way that is quite impossible at present. I take it for granted, for example, that free, high quality, semantically-aware XML-based mathematics will become available during 1998. This leads directly to the interoperability of MathML and any other XML-based language.
the chemical community was briefly able to achieve during the period 1996-1997 with the "old style" web browser and a plug-in such as Chime was just a temporary phenomenon. Arguably, one might look back to that period as a golden, simple one where many people could be molecularly enfranchised with little effort. We may never see its like again.
Do you want the 5-minute argument or the full half-hour :-)? Seriously, I'd like to call it the dawn of the golden age. Molecular plugins/helpers had the same breathtaking effect as HTML had for text/graphics. XML is the next logical step forward for both. It requires more engineering, but it can deliver a great deal more. P. Peter Murray-Rust (PeterMR, http://www.nottingham.ac.uk/~pazpmr) Director, VSMS (Virtual School of Molecular Sciences) Pharmaceutical Sciences, Nottingham Univ. NG7 2RD, UK; T: 44-115-9515087 F: 5110 http://www.vsms.nottingham.ac.uk/ chemweb: A list for Chemical Applications of the Internet. To post to list: mailto:chemweb@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/chemweb/ To (un)subscribe, mailto:majordomo@ic.ac.uk the following message; (un)subscribe chemweb List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)
participants (2)
- 
                
                peter murray rust
- 
                
                Rzepa, Henry