| [Thread Prev] | [Thread Next] | [Thread Index] | [Date Prev] | [Date Next] | [Date Index] |
On Fri, 1 Apr 2005, Mike Everhart wrote: > Date: Fri, 1 Apr 2005 11:43:56 -0600 > From: Mike Everhart <mike@oceansofkansas.com> > Reply-To: paleonet@nhm.ac.uk > To: paleonet@nhm.ac.uk > Subject: RE: paleonet Paleo Portal > > > As a contributor of several to the photos posted on Paleo Portal, I > would certainly agree with his concerns. It pains me a bit to see my > large, detailed photographs reduced to a less than useful size. I know > what costs are associated with maintenance of a photo-intensive web site > and don't think that what I am paying personally is beyond the budget of > a much larger organization. Dear Mike, Dear others who think about such a website, I suspect that the main point was not the resolution or the disk space. Formulated simplified and in this way not fully correct: When files are clicked, they are squeezed into the internet. When two users click, the two files are squeezed synchronously and so forth. Universities have often 155 MBit or 600+ MBit (inhouse sometimes Gbit). There it (factually) does not matter. When 30 Paleontologists around the world click the same uploaded trilobite (high res) the server might appear down if it is not a computer center of a university (in above sense) in the background. Thus I propose to think about two alternatives: 1) Each medium-res photograph might be connected with a link to an ftp connection that transmits the high res. Ftp works in a different way than the, formulated extremely compact, "normal clicking". 2) The large commercial providers, T-Online in Germany, in the US there is also a subsidary, offer for quite low rates also "business hosting". This means: High transfer rates, ample space and always up to date background technology. In case of load fluctuations they adjust without the users noting anything. The "business details" are revealed on request. An NSF-supported site which aims at becoming a high-end reference portal must, also at this early time, consider a much larger audience and thus much larger traffic than a specialized site, comparable to Porsche (small high quality company, small logistics) see as example Oceans of Kansas and Mercedes (also high quality, but worldwide availability of all spare parts in large quantities), see the NSF portal. Above applies especially to strong load fluctions, for example after a Nature article on feathered dinosaurs (tsunami load) comparable to normal load. Possibly they might consider a demand-dependent (automated) mirroring of _those_ files that require temporarily full-strenghth capabilities (such as an OS/390). As IBM for example made the Eremitage available and as they consistently advertise their business on demand line, they might also in the US have a respective branch. On the tarriffs of IBM I cannot comment. Thus I think they are apologized. After measuring the traffic etc. I think they will adjust to what is needed, technologically possibly to a heterogenous solution in above sense. > > My two cents... ..., with a polite smile: 0.1 added. > > Regards, > > Mike Everhart Best regards, Peter ********************************************************************** Dr. Peter P. Smolka University Muenster Geological Institute Corrensstr. 24 D-48149 Muenster Tel.: +49/251/833-3989 +49/2533/4401 Fax: +49/251/833-3989 +49/2533/4401 E-Mail: smolka@uni-muenster.de E-Mail: PSmolka@T-Online.de **********************************************************************
Partial index: