Interest In NTI/48.2 Project
Academic Computing Services , University of Sheffield
Stuart Herbert (S.Herbert@Sheffield.ac.uk)Document copyright ©. All rights reserved.
Abstract
JISC, through its New Technologies Initiative, has funded the
University of Sheffield for one year to supply and support UNIX
batch processing systems.
Contents
Click here for a plain-text version of this paper. Click here for a copy of this document in Microsoft RTF format, suitable for printing (if available).
Introduction
About This Paper
This paper presents tabulated information obtained from the logs
kept by our ftp and World Wide Web servers regarding our NQS
documentation and files.
UK Academia
Introduction
UK Academia is, by nature of our funding, considered to be our main,
and most important, customers.
Table 1 : FTP/WWW/Mailbase By UK Academic Site
The following table shows :
- which UK Academic sites have downloaded files from our NQS
Archive
- which UK Academic sites have downloaded documents from our
NQS WWW Support Service
- which UK Academic sites originally expressed an interest in
our product (in response to our original questionaire)
- which UK Academic sites have subscribed to one (or more) of our
Mailbase lists
> -----------------------------------------------------------------
> Site | FTP | WWW | Interested | Mailbase
> -----------------------------------------------------------------
> aber.ac.uk | 36 | | No Reply |
> afrc.ac.uk | | 1 | No Reply |
> anglia.ac.uk | 1 | 11 | No Reply |
> aston.ac.uk | 8 | | No Reply |
> bath.ac.uk | | 17 | No Reply |
> bbk.ac.uk | | 6 | No Reply |
> bham.ac.uk | | 49 | Yes | Automatic
> brad.ac.uk | 4 | 9 | No Reply |
> bris.ac.uk | 1 | 54 | No Reply | Yes
> brookes.ac.uk | 3 | 2 | Yes | Automatic
> brunel.ac.uk | 11 | 31 | Yes | Automatic
> cam.ac.uk | | 8 | No Reply |
> cf.ac.uk | 19 | 43 | Yes | Automatic
> cranfield.ac.uk | 10 | 19 | Yes | Automatic
> dl.ac.uk | 21 | 82 | Yes | Automatic
> dur.ac.uk | | 14 | Yes | Automatic
> ed.ac.uk | 12 | 113 | No Reply |
> essex.ac.uk | 3 | 26 | No Reply |
> ex.ac.uk | | 39 | Yes | Automatic
> gla.ac.uk | 13 | 144 | Yes | Automatic
> glam.ac.uk | | 11 | No Reply |
> gre.ac.uk | 2 | 6 | No Reply |
> hensa.ac.uk | 5 | 126 | No Reply |
> hud.ac.uk | | 4 | No Reply |
> hull.ac.uk | | 23 | Yes | Automatic
> hw.ac.uk | 1 | 20 | No Reply | Yes
> ic.ac.uk | 21 | 48 | No Reply |
> kcl.ac.uk | 1 | 4 | No Reply | Yes
> keele.ac.uk | | 21 | Yes | Automatic
> lancaster.ac.uk | | | Yes | Automatic
> le.ac.uk | | 23 | Yes | Automatic
> leeds.ac.uk | 11 | 12 | Yes | Automatic
> liv.ac.uk | | 35 | Yes | Automatic
> lmu.ac.uk | | 17 | No Reply |
> lut.ac.uk | | 5 | No Reply |
> man.ac.uk | | 7 | No Reply |
> mcc.ac.uk | 2 | 49 | Yes | Automatic
> mmu.ac.uk | 11 | 15 | No Reply |
> ncl.ac.uk | | 52 | No Reply |
> nene.ac.uk | | 2 | No |
> nott.ac.uk | | 11 | No Reply |
> nottingham.ac.uk | | 3 | No Reply |
> ox.ac.uk | 19 | 62 | No Reply |
> paisley.ac.uk | | 8 | No Reply |
> qmw.ac.uk | 1 | 37 | No Reply |
> qub.ac.uk | | 22 | No Reply | Yes
> rdg.ac.uk | 19 | 19 | Yes | Automatic
> rl.ac.uk | | 64 | Yes | Automatic
> salford.ac.uk | | 24 | Yes | Automatic
> sbu.ac.uk | | 2 | No Reply |
> shef.ac.uk | 40 | 943 | Yes | Automatic
> soton.ac.uk | 1 | 123 | Yes | Automatic
> st-and.ac.uk | | 3 | Yes | Automatic
> stir.ac.uk | 35 | 3 | No Reply | Yes
> strath.ac.uk | 37 | 91 | Yes | Automatic
> surrey.ac.uk | | 2 | No Reply | Yes
> swan.ac.uk | | 14 | No Reply |
> ucl.ac.uk | | 3 | No Reply |
> uea.ac.uk | 16 | 74 | Yes | Automatic
> ukc.ac.uk | 2 | | Yes | Automatic
> ulcc.ac.uk | | 5 | Yes | Automatic
> ulst.ac.uk | 1 | | Yes | Automatic
> unn.ac.uk | 3 | 21 | No Reply |
> warwick.ac.uk | | 2 | No Reply |
> wmin.ac.uk | | 4 | No |
> york.ac.uk | | 8 | No Reply |
> -----------------------------------------------------------------
Table 2 : Summary Of UK Academic Interest
The following table shows a breakdown of the figures in Table 1.
Percentages shown are a percentage of all UK Academic Sites listed
in Table 1, except for `files downloaded by ftp' and `documents
downloaded by WWW', which are percentages of all downloads
world-wide.
> -----------------------------------------------------------------
> Total Percentage
> -----------------------------------------------------------------
> UK Academic Sites In Table 1 : 66 Unknown
> -----------------------------------------------------------------
> Sites Subscribed To Mailbase Lists : 33 50%
> who have downloaded files : 19 29%
> who have downloaded documents : 30 45%
> who expressed an interest in
> response to questionaire : 27 41%
> -----------------------------------------------------------------
> Sites Who Expressed An Interest
> In Response To Questionaire : 27 41%
> who have downloaded files : 15 23%
> who have downloaded documents : 24 36%
> -----------------------------------------------------------------
> Sites Who Have Downloaded Files : 32 48%
> who have downloaded documents : 28 42%
> who are subscribed to Mailbase : 19 29%
> -----------------------------------------------------------------
> Sites Who Have Downloaded Documents : 61 92%
> who are subscribed to Mailbase : 30 45%
> who have downloaded files : 28 42%
> -----------------------------------------------------------------
> Files Downloaded By FTP : 369 9%
> Documemts downloaded by WWW : 2664 42%
> -----------------------------------------------------------------
Conclusions
From this data, we have reached the following conclusions :
- At least six UK Academia sites are using our product.
This is based on sites which have downloaded files, and then
subscribed themselves to the Mailbase lists. This figure does
not include any other evidence.
- We believe that at least ten UK Academia sites are using our
product.
This is based on the evidence above, that we are using our
product on-site, and correspondance with three other sites (aber,
cf, leeds) indicating that they have installed our product.
- We believe that it is extremely likely that between nineteen and
thirty-two UK Academic sites are using our product.
Thirty-two sites have downloaded files.
50% of these sites showed interest in response to our
questionaire - we consider it highly unlikely that they would
have downloaded the product but not be using it.
Twenty-eight sites have downloaded documentation and files - we
consider it highly unlikely that many of these sites would then
have rejected using the product.
Nineteen of these sites are subscribed to our Mailbase lists. If
they believed the product to be unsuitable for their use, one
would expect these sites to have unsubscribed from our Mailbase
lists.
We feel that the figure of twenty-seven sites using our product
is a reasonable estimate from the data available.
- No more than thirty-two UK Academic sites are using our product.
This is based on the number of sites who have downloaded files.
UK Non-Academic
Table 3 : FTP/WWW/Mailbase By UK Non-Academic
> -----------------------------------------------------------------
> Site | FTP | WWW | Mailbase
> -----------------------------------------------------------------
> bt.co.uk | | 2 | No
> collins.co.uk | | 5 | No
> demon.co.uk | | 3 | No
> eamwf.co.uk | | 26 | No
> icl.co.uk | | 6 | No
> intera.co.uk | | 2 | No
> mtd.co.uk | 1 | 1 | No
> meto.govt.uk | 4 | | No
> nag.co.uk | | 16 | No
> intmarkt.demon.co.uk | 4 | | No
> mondrian.demon.co.uk | 1 | | No
> sbil.co.uk | | 6 | No
> -----------------------------------------------------------------
> Total | 12 | 10 | 67 | 0
> -----------------------------------------------------------------
Conclusions
- UK Non-Academic sites have little knowledge of our product.
World-Wide
Table 4 : FTP/WWW/Mailbase By World-Wide Usage
> -----------------------------------------------------------------
> Country | FTP | FTP % | WWW | WWW % | Lists | Lists %
> -----------------------------------------------------------------
> Austria | 4 | 0.1% | 31 | 0.5% | 0 | 0.0%
> Australia | 52 | 1.3% | 104 | 1.6% | 4 | 3.4%
> Belgium | 24 | 0.6% | 13 | 0.2% | 0 | 0.0%
> Brazil | 0 | 0.0% | 2 | 0.0% | 0 | 0.0%
> Canada | 31 | 0.8% | 158 | 2.5% | 2 | 1.7%
> Denmark | 32 | 0.8% | 2 | 0.0% | 0 | 0.0%
> Finland | 29 | 0.7% | 34 | 0.5% | 0 | 0.0%
> France | 42 | 1.0% | 120 | 1.9% | 0 | 0.0%
> Hungary | 0 | 0.0% | 12 | 0.2% | 0 | 0.0%
> Germany | 276 | 6.8% | 501 | 7.9% | 14 | 11.9%
> Hong Kong | 32 | 0.8% | 11 | 0.2% | 0 | 0.0%
> Iceland | 0 | 0.0% | 5 | 0.1% | 0 | 0.0%
> Indonesia | 0 | 0.0% | 4 | 0.1% | 0 | 0.0%
> Ireland | 0 | 0.0% | 1 | 0.0% | 0 | 0.0%
> Israel | 6 | 0.1% | 13 | 0.2% | 0 | 0.0%
> Italy | 19 | 0.5% | 45 | 0.7% | 0 | 0.0%
> Japan | 35 | 0.9% | 56 | 0.9% | 0 | 0.0%
> Malaysia | 0 | 0.0% | 1 | 0.0% | 3 | 2.5%
> Netherlands | 19 | 0.5% | 81 | 1.3% | 0 | 0.0%
> Norway | 5 | 0.1% | 5 | 0.1% | 0 | 0.0%
> Poland | 3 | 0.1% | 0 | 0.0% | 0 | 0.0%
> Singapore | 2 | 0.1% | 27 | 0.4% | 0 | 0.0%
> Spain | 49 | 1.2% | 41 | 0.6% | 3 | 2.5%
> South Africa | 0 | 0.0% | 18 | 0.3% | 0 | 0.0%
> South Korea | 21 | 0.5% | 47 | 0.7% | 0 | 0.0%
> Soviet Union | 0 | 0.0% | 4 | 0.1% | 0 | 0.0%
> Sweden | 286 | 7.0% | 187 | 2.9% | 3 | 2.5%
> Switzerland | 21 | 0.5% | 135 | 2.1% | 4 | 3.4%
> UK Commercial| 10 | 0.3% | 67 | 1.1% | 0 | 0.0%
> UK Academic | 369 | 9.0% | 2664 | 41.9% | 61 | 51.7%
> Unknown | 89 | 2.2% | 532 | 8.4% | 0 | 0.0%
> US Commercial| 173 | 4.2% | 565 | 8.9% | 16 | 13.6%
> US Academic | 2374 | 58.2% | 556 | 8.7% | 5 | 4.2%
> US Government| 65 | 1.6% | 297 | 4.7% | 3 | 2.5%
> US Military | 12 | 0.3% | 24 | 0.4% | 0 | 0.0%
> -----------------------------------------------------------------
> Total | 35 | 4080 | | 6363 | | 118 |
> -----------------------------------------------------------------
- The US site sunsite.unc.edu provides a US mirror of our NQS
Archive (and this mirror is mirrored around the world), making it
impossible to predict just how many copies of NQS are installed
world-wide.
Conclusions
- The software product of this project is of significant interest
world-wide.
This suggests that the software product would be successful if
sold commercially, if priced agressively to compete with the
established market leader, Cray NQE.
This also suggests that the NTSC could use this project in, say,
a portfolio of successes when negotiating with JISC for increased
resources for the NTI ...
- The WWW-Support product of this project is of significant
interest to UK Academia, but is not of significant interest
world-wide.
This is probably due to the very poor links between SuperJANET
and the Internet proper, making WWW access from outside the UK
too slow for interactive use (ftp at least is something one can
walk away from, or schedule to do during not-so-busy times).
Once Sheffield-NQS is released, all WWW documentation will be
provided with each release, in an attempt to ease the situation.
- The Email-Support product of this project is of interest to UK
Academia, but is not of interest world-wide.
This may be due to Mailbase's policy of mainly-UK use - typically,
one finds mailing lists around the world to be highly popular.
|