There were 366 responses to the survey, very similar to the 364 received last year. Below is a plot showing the response rate for the past 5 years. In 2002, we did not ask if users were employees of STScI.

We solicited responses from all subscribers to the Archive Newsletter, those who had a GALEX CASJobs account, and to all those who had retrieved data from DADS in the previous 28 months. We also placed an invitation to participate in the survey on the MAST web pages.

With questions 1 through 8, we asked questions to characterize MAST usage and to determine the platforms and browsers MAST users are utilizing. The remaining questions attempt to solicit input concerning various tools and services MAST has either implemented or may plan to implement in the future. We provided several places for responders to enter comments and suggestions. We have selected a subset of the comments and included them in this summary. If several people had the same or similar response, we have only included one of them. In some cases, similar comments were included in two different comment fields. We would like to respond to some of the suggestions and comments. That response has been compiled in a separate web page. In some respects we find the comments the most useful part of the survey. We would really like to get these throughout the year as you discover needs or have questions. You may send suggestions to the archive help desk at archive@stsci.edu or call at 410-338-4547 between 9:00 AM and 5:00 PM eastern time. You may also wish to use the MAST Suggestion Box.

1. How often have you used MAST in the past 12 months?

3. What missions did you access? (multiple choices permitted)

4. Which operating system do you primarily use to access MAST?

5. Which browser do you primarily use for your work?

6. Would you be able to utilize search/data access tool that only operates in Firefox?

273 Yes
72 No

Some comments concerning this question:

• I would be able to use a search tool that operates in firefox, but in general I think it's best not to require a tool that requires the use of a specific browser.

• I think it would be a disservice to the community to assume that one would only serve a single browser, e.g FireFox as one of the questions suggested you might be considering. This in my opinion is a recipe for problems, since even if FireFox is dominant today, there is no guarantee that it will be tomorrow.

• At the moment I am using a LINUX system with Firefox implemented and I could use this to use MAST facilities. It is anticipated, however, that Mac OS-X will become the operating system of choice over the next few months and Safari the browser to be used; this is the reason why I have answered Question 6 in the way I have.

  

  

7. For what reasons do you use MAST? (check all that apply)

Not surprisingly scientific research is the most common reason for using MAST. There are a growing number of amateur astronomers who use the archive.

8. What is your primary and most common reason for using MAST?

9. Which of the following have you used in the past 12 months for learning about new features or developments at MAST?

Comments about learning about new features or developments at MAST:

• Four responses indicated e-mail notification was satisfactory.

• I believe communication is as effective as it needs to be. I have never concerned myself with new developments in MAST as I understand the present system provides all that is required.

10. In what ways do you use MAST's Community Contributed High Level Science Products (multiple choices permitted)

Comments:

• Provide better quick-look data

• I have made no use of High Level Science Products up to now, but will be doing so in future.

11. The GALEX team has been working on some upgrades. Please enter your priority for these 5 projects.

We present two views of the responses to this question. The first plot shows all responses to this question. However, we noticed that people who had not used GALEX data had responded to this question. The second plot shows the responses from those who had indicated use of GALEX data. While the responses were similar, one could see that those who had used GALEX data were more enthusiastic about some options (such as the customized downloads) than the general group.

Comments:

• image cutouts of user-selected size and sky location with co-addition from all relevant tiles

• Easy cross-correlation with SDSS, 2MASS, etc would be *very* useful.

• What does google-like maps mean for astronomy images?

• make it work like dss in two bands ie selectable regions instead of whole images

• My principal interest would be to know, after MAST serves me a/some GALEX file(s) and I choose one and see a thmb sketch I like, *which* of the many downloadable images I should get. Knowing nothing detailed about GALEX formats, I would like to know

12. Which of the following services have you used?

Comments:

• I use the MAST/Aladin connection a LOT when I use it (often during ground-based observing runs)

• Used Aladin for HST proposals.

• Vizier Catalog tool doesn't work well.

13. MAST and STScI are currently pursuing plans for the Hubble Legacy Archive (HLA) with fast data access through MAST and VO tools. Proposed new features include immediate access to calibrated HST data, improved astrometry, footprint services giving accurate image boundaries, image cutout services, and object catalogs. Does this sound useful to you?

Are there additional capabilities you would like to see?

We list most of the comments below (some duplicate comments are not included). Our href="/surveyresults/2007/response2007.html"> response is on a separate page.

• I'm looking forward to the footprint service etc. from the HLA. That's been a long time in coming.

• Am very much looking forward to HLA and hope it will make intercomparison of various space- and ground-based images more easy. The need for such intercomparison is important part of my daily research. Thank you for all your efforts!

• A secure way for HLA to work for datasets which are still proprietary (i.e. produce nice products for the PI only). Also, some way to produce useful products for moving target (Solar System) datasets, which have unique features and desired products.

• Deliver routinely drizzle/mosaicked WFPC2 data (as well as ACS) so as to allow quick analysis using tools so that users (especially students) don't have to do that layer with STSDAS possibly requiring an unfamiliar machine and OS.

• A capability most useful to me would be the optional capability of searching archives of major ground-based observatories from within MAST.

• Yes. When I perform e.g. a STIS search, and select user-selectable fields to be returned that are not default (e.g. start wave, end wave, resolution etc), these should be retained when I use the 'back' button of my browser, in order to facilitate searching for multiple targets.

• Is there a conversion tool for coordinate input, especially for batch processing? Many 'classic' astronomers use sexigesimal format, but newer catalogs and search engines use/require decimal degree input. Or can you point me to a good conversion tool... preferrably web based with file input (used for research and software testing).
• Search for images at any of an extended list of positions

• co-added, WFPC2 images, with CRs cleaned, and accompanying products for that co-added image: net exposure time map, estimated noise map

• Immediate access to non-drizzled calibrated HST data is exactly what the doctor ordered, and I see you have that covered above.

• ACS associations, like the WFPC2 ones.

• access to calibrated HST data is fast enough already

• For each data set (in particular HST), quoting limiting 3 sigma magnitudes for point sources/extended sources - in addition to the exposure times - would be a useful feature.

• For spectroscopic data, it'll be usefull to get a simple image with the configuration overplotted on the object

• correlate with (planned) footprints of future surveys/missions

• 1) coaddition of largely (say more than 70 %) overlapping images, even if taken in different orbits. 2) search for exposures longer than a given staked (i.e. integrated over all exposures) exposure time (e.g. sky locations with F850 texp>10000s)

• access to calibrated spectra in ascii format (ascii tables with wave, flux, error for all the HST spectra e.g. FOS, GHRS, STIS,...)

• Most useful: improved astrometry, calibrated images Less useful: object catalogs. Most desirable: Ability to get GALEX and CHANDRA images on same quick-look format and scale, for immediate comparison, blinking, and/or printout.

• Angular sizes of images be given or alternately it should be possible to quickly estimate the sizes from (preview)images even though it may be only a good approximation.

• Final Published Catalogs of data (Authors should make the actual ascii catalogs of their photometry or spectra available)

• Searches on flats and darks etc for HST, particularly NICMOS

• The ability to download FITS Files converted to 24 bit RGB and both 24 bit and 16 bit grayscale

• If I'm getting data for science purposes I would reduce the data myself anyways and being able to access calibrated data would not be terribly useful (useful for a better quick-look).

• 1) Ability to query for and retrieve exposure level data with root data set names, not "lumped" into associataions via the MAST web intrface. 2) Ability to query for (in particular) NICMOS calibration and science data qualified by SAMPSEQ for multiaccum images (i.e., to qualify and retrieve relevant [dark] calibration data via web interface 3) Ability to search for (NICMOS in particular) LAMP ON frames via web inteerface.

• Spectral querying capabilities.

• Complete blue plate (and hydrogen-alpha) coverage for DSS images. Ideally false colour images that combine red, blue, h-alpha and infra-red.

• exoplanets search

• Far better than IRSA. Not as user friendly as NED.
• Since I work on nearby galaxies of sometimes large angular size, I use mainly NED, and every now and then want to know whether there are new images of my objects of interest available. I find that the MAST interface improved over time, but NED is still far more useful, since it is more oriented to a 2D picture of the sky, and not to a list of point sources. The above mentioned improvements should greatly improve the usefulness of MAST for my purposes.

• Incorporate NED and SIMBAD results to the targets searched for.

• similar to others, would like to see a little more support for multi-object searching.

• HEASARC has a free-text 'intelligent' search query form that I find quite useful.

• I also use CADC and ESO. Turn-around times for requests are typically somewhat faster at those two archives. I find it pleasant if series of observations are returned as one 'association'.

• Download times for HST data are SLOW (~1 day) compared to Chandra and Spitzer. Spitzer datasets are of course much smaller, but Chandra datasets are big, and they still stage w/in ~20 minutes of the request.

• Much easier to use than the PDS!

• CADC is amateurish by comparison.

• Seems fairly straight-forward for doing simple target searches. The links to proposals and papers, etc, are a nice touch.

• Do you have a batch processing/searching capability yet?

• I find MAST to have aslightly higher learning curve and the web-based interface not always intuitive. However I am a fond user of the HST search tool and I would like to attend a workshop at the next AAS. Compared to Spot, it is slightly less user-friendly but both have problems.

• Easier to use than other data centers

• I find MAST very frustrating to use for moving targets. It is easily searchable for coordinates and objects with single designations, but not for objects with different coordaintes depending on when they are observed. Since newly discovered solar system objects go through a designation process they are often not easily searchable by name since you need to know all the possible names people could have given it when the observations were taken. Maybe there is some way to link to jpl ephemeries or some other orbit reference for moving objects, just a thought.

• MAST is already extremely useful, although when searching for info on a particular object, I still find SIMBAD more informative

• CasJobs-like interface + limited personal storage for users would be very useful and much better than current VO tools.

• My preferred IUE spectra in ASCII format is hard to find, sometimes.

• I find the MAST web site very useful, but not well organized and not so easy to navigate and find your way around.

• MAST is OK, compared to other data centers. It's a little harder to find things on the MAST site compared to HEASARC or Chandra. It's dealing with more missions than Spitzer. I find that I have to know something exists before I find it, so I'm not using many things mentioned above. I've not heard of DataScope or Open SkyQuery, although I may have well used them without knowing their code names. It would be nice to see a simpler, cleaner entry to the site.

• Except for the immediate access to calibrated HST data (non-drizzled) yours is the best site to navigate. Even if you only offer the calibrated HST data binned (al la preview) in a certain mode, that would be a vast improvement.

• I have found that accessing HST data is more complicated than accessing (e.g.) Chandra data. This may be because HST has more security safeguards in place. If so, are they necessary?

• I have tried other data centers, and did not like them. Mast is far better.

• Spitzer's data retrieval program, Leopard, has a feature that overlays outlines of spectral slits and image fields of view from a selected observing program on sky survey images. A tool like this would be very useful in MAST. For example, I'd like a quick way to see what areas of a nearby galaxy were observed during a specific program.

• Simple and powerful; especially useful are image previewing for archive data. Much easier than e.g. XMM-OM archive at XSA.

• easy to use, nice search and navigation, but in the last week the staging disk is always full and I am not allowd to give my password for local security reasons for sftp -> I'd like more staging capacity, possibly some reserved space for data sets < 50 MB (If I need really large amounts of data if is worth taking more trouble, but if I just need quickly a callibration spectrum or a secondd short 10 MB observation I have no alternative to the staging disk)

• Being able to browse the sky-position of datasets (e.g. SDSS Explore/Navigate) and select tiles, observations or individual objects within different datasets within MAST would be very useful. A link from each mission page to its own browse window, perhaps.

• I did not realize that I am going through MAST when I access and download HST data. This has worked well for me. I have found access to GALEX data more difficult, in part because I know little about it. What I have missed the most is *finding* MAST access to GALEX data for some specific galaxy, be it from NED, via the GALEX home page, or otherwise. Somehow, the fact that MAST serves GALEX images seems hidden. Better crosslinks from other sites (esp. NED) would help. Example: Am looking up my favorite galaxy on NED. Don't find GALEX images there. Have to know that I can click on some MAST link near bottom of NED display. But even that link does NOT bring up any info about my current-interest galaxy. Have to go somewhere to upper left and click on some 'Search form' link or the like. Cumbersome, should be easy to do better than that.

How can we do a better job supporting your archive needs?

• Expand the usual 10 search results per instrument to include all available (or have an option for such). It's tiring to have 13 results and have to go find the last three separately.
• You used to have a feature where the DSS overlay was visible with WFPC2 archive data- can't seem to find it now.

• MAST is the most useful because it feaures so many missions. It is quite slow however, and very often the 'Reference' information associated with a target search is incorrect. I find this to be particularly true for FUSE.

• ESO data archive center did not allow non-member country astronomers from accessing their data (as per my personal experience some years ago). Compared to that MAST is doing wonderful service to the poor astronomers of the world at large. Kudos to MAST and their providers!!

• At present, I do not find it straightforward to download data to my laptop as is is possible to do with Spitzer software tools. Thus I end up with a two stage process, downloading the Spitzer data to my desktop and then transferring it to my laptop where I actually do most of my work.

• I predominantly use vizier, irsa/ipac GATOR service, galex casjobs and sdss casjobs. Always useful are http get cone searches that can take decimal degree arguments that also return flat datafiles. vizier, irsa/ipac and sdss can take these and return these (albeit comma delimited ascii file in the case of the latter, whilst vizier often has embedded html code that still needs stripping out of returned ascii data). Currently, unless I'm mistaken, MAST only does this through cross correlation features. It's be nice to return simple GALEX information, position, mags, errors on mags, and probably some flags, in a flat ascii form in this way. Then they can either be plotted up in graphical star charting interfaces, or can be matched against other lists, either locally, or via upping them against GATOR or VizieR list batch processing functions. Granted something similar can be done using casjobs, but it does seem like overkill for something this simple.

• SuperCOSMOS has better blue plate coverage for southern hemisphere DSS images as well as excellent hydrogen-alpha - I don't understand why MAST can't have all this data in one site.

• as an occasional user, when looking for a specific proposal id, i have to wade through a whole bunch of generic search options and know how to set specific options. e.g. proposal 10799 search option 1: generic HST search (http://archive.stsci.edu/hst/search.php) - proposal ID one box among many - have to know to set Maximum Records to > 105 Search option 2: HST Abstract search (http://archive.stsci.edu/hst/abstract.html) http://archive.stsci.edu/cgi-bin/proposal_search?id=10799&mission=hst - only returns the first 100, with no way to see the rest

• I still find some of the links confusing... expecially what is the difference between MAST FUSE Science and MAST FUSE Exposures???? Usually I cannot retrieve my very own data when it is still restricted without sending a special email to MAST... how come you all can not remember my login and p/w? :)

• Faster FTP.

• make data available quicker on ftp

• VO stuff is a great concept but tools and integration are still implemented (across the board) poorly and are still not useful for scientific research.

• PLEASE PLEASE PLEASE improve the keywords search for HST data. I would like to have wildcards, loose aliases, etc. The last time that I checked you had to hit the keywords exactly to find what you're looking for.

• make it easier to select HST files after the initial search. e.g. limit the exposure time and filter of all the files found. Also: search by combined exposure time. A google-like map with all the exposures near a given object would be good.

• It has already been addressed in one of the questions, but I'd like to point out that having some cross-correlation showing what data is out there for each object in 'the most popular' databases (e.g. SDSS, 2MASS , XMM, FUSE etc). Also, although it is vitally important that 'raw' data is always provided, it is becoming more and more necessary for non-specialists to have access to well calibrated data for 'quick and dirty' comparisons with other regions of the e-m spectrum taken with instruments that the researcher may not be familiar with. So, more fully reduced and claibrated data would be ideal.

• I would be able to use a search tool that operates in firefox, but in general I think it's best not to require a tool that requires the use of a specific browser.

• Assuming that there is no feature now of which I am aware, I wonder about the possibility of having a keyword word/phrase that can be attached to a retrievals request and that gets attached to the archive correspondence about the retrieval status. (i.e. "NGC 602"; "NICMOS Parallel ANY"; "Prop 10204", etc.) Secondly, it would be nice to be able to view a summary of recent retrievals and where the output was sent. Lastly, it would be nice to have a permanent time of deletion on the staging disk or if that time floats, a 12 hour reminder that the data is being delete

• standardized data retrieval process/options for all missions for imaging and spectral data.

• A primary web page with a list of all services (some sevices are well hidden). Access to the GALEX images (not via query). Adding 2MASS and GALEX images to the POSS service.

• shorter times between requests for data and delivery

• More preview images of ACS data.

• these are very basic suggestions, which i would hope could be addressed before any of the grand high-level features discussed in this survey: * HST search form cannot display all records. for example, i got this message at the top of the search results page: '100 rows displayed, but 2397 are available.' since MAST can't display more than 1500 records, a large portion of the search results are inaccessible. solutions: a 'next' button, or a 'show all records' option. * HST search form could have an option to sort results by any column (date, proposal ID, etc) * HST search form 'mark' buttons and 'submit' button are too close together; i have hit 'mark' buttons by accident instead of submit, and lost all the marks i had made on the form.

• I would like to have an easy link to get complementary information of each HST image. I'm thinking in things such as decontamination dates near the date of UV images, electronic gain value, and alike.

• Objects in the IUE archive can be easily selected through an user friendly facility based on an Object classification code (e.g. one can see all available IUE spectra of Supernovae entering code 56 in the proper box ) . I am not aware of a similar facility for spectra or images taken with other instruments. I think it would be useful to have EASY access to all data belonging to a class of objects.

• I like most 'give me back the fits image, reduced as good as possible, the zero point and pixel size, that I will make my own catalog, that I will cross-match to other data in the way I prefer'. I hate 'advanced products', for which it is unclear which, of the many files, is the only important one for my own research, how to export them in my favorite software, what is the meaning of the different files, and columns inside the files, etc. or when all the above become clear *after* reading hundreds long documents.

• You guys are doing a great job, thanks a lot. As I mentioned, I would glad to see spectra downloadable in ascii format such as ascii tables with wavelength, flux, error, as the ones that already exist for the IUE archive. Since I use mainly HST and FUSE I would be happy to see this implemented for HST spectra and FUSE spectra (which are much more difficult to combine than HST). I would even be ready to participate in such a project that would put HLSP (high level science product) spectra (callibrated, coadded, etc..) for FUSE

• Aperture Overlays would be very useful. A useful tool would be to have a DSS image appear during a search, with spectrogrpah aperture overlays at the appropriate coordinates. If this would be too busy, then some sort of interactive tool that would allow the user to specify overlays would be useful. Such tools are available through Spitzer-Leopard and are VERY useful. Also, the format of the Coperinicus data makes it text to useless. Something more akin to the ascii format that IUE is returned in would be much more useful.

• I have used in the past some support services like clarifying doubts etc., and found pretty good. A general wish is that it should be possible to reduce/analyze multifarious archival data with just one common package like IRAF. Having new and different packages for each new mission by each group of nations/institutions is very bothersome (for the user community) and is restricting the number of publications that can come out of all archival data. It can also save lot of cost, time and energy for the archive support organizations.

• GSC2.3.2 is now available via vizier, which is useful, but the long promised decimal degree http get interrogation, as opposed to the current sexagesimal only system, is still not available. Therefore vizier will, at least in my case, become first choice for retrieving this data, rather than via MAST.

• I just appreciate having access to these images on an indivdual level, Thank you for your part in making them accessible.

• A write-up (or howto) on the best ways to manipulate HST data in a non-IRAF environment would gather great interest in the non-professional community.

• it would be nice to see every image taken with Hubble in full resolution and fully reduced in the Aladin sky atlas, and more regular and faster availability of preview images,as soon as they are free of propriety. I am a very interested amateur astronomer, who likes to see more of what you do, would be nice if i as an amateur can offer proposals for Hubble observations, since Hubble is the ultimate tool for looking at the universe. Keep up the good work please!

• A simpler interface, more straightforward help files and general googlablity would be useful. I had to get a list of HST calibration files. And my various searches of 'MAST download calibration files' turned up nothing useful. I had to go to several help pages before I found the one that told how to get files rather than datasets.

• Updates to the OTFR pipeline sometimes result in substantial changes to the data products retrieved from MAST. Under some circumstances a MAST user may want to register to receive a notice when the OTFR pipeline is changed in a way that would affect the data they have retrieved. This may be a non-trivial feature to add, and perhaps the pipeline is at a stable point now where such notices are not really needed. Something to consider.

• In your front search page (http://archive.stsci.edu/), please make one of the resolver options 'Don't Resolve'! Currently the only two options are 'SIMBAD' and 'NED' but neither works for some things (e.g., solar system objects).

• Have sftp available for downloading data. Update some explanatory paragraphs to be a little more detailed.