← Previous Contents

News

XTide 2.11

XTide 2.11 fixes a few bugs, adds GPS integration (still somewhat experimental) and improves compatibility with the latest window managers and libraries.  Get it from http://www.flaterco.com/xtide/files.html.

Data quality

For several years there have been two tide stations for which the meridians were set wrong in the data as incoming (ostensibly UTC data were actually calibrated to local time).  However, a diligent user recently found two stations in California that newly developed this problem between December 2008 and December 2009.  As always, please use caution and report such problems when they are found.

Mailing list

There is a manually-maintained, announce-only mailing list to which I send notices of each update to XTide or the harmonics data.  You can request to join by e-mailing me at dave@flaterco.com.

Please note:  If an announcement to you gets bounced by your mail server, you fall off the list.  Some long-time contributors have fallen off because, unbeknownst to them, their ISPs subscribed to a spam blackholing service that blackholed all pair.com-hosted domains after one of them sent out spam.

Roadmap to future development

Things worth fixing / TO-DO list

Data maintenance

I refresh the data for the U.S. once a year from the National Ocean Service web site.  This is a lot of work and takes up basically all of the time and energy that I can devote to data maintenance.

The coverage of other countries suffered a major blow in 2001 when the UKHO claimed to own the copyright on tons of old tide data that I had painstakingly processed for XTide.  With the help of some diligent volunteers, I eventually managed to source new data for the U.K., Germany [*], Netherlands, and Canada.  Since then, a slowly growing group of countries has made harmonic constants or water level records available on the web.  Usually, the legal terms attached to those data are written to grant free license only in the case of direct personal or academic use.  The XTide database is not exactly that, since the data or a derivative work would be redistributed to untracked masses of users, possibly through commercial channels such as packaged Linux distributions.  But even if the legalities were cleared up, the annual job of pulling data off of a web site, developing ad hoc conversion tools that might never work again (even for the same web site one year later), cleaning up the converted data, and integrating them with the XTide database is too much for me when multiplied by N different countries.

I encourage those tide authorities that are now making harmonic constants publicly available on their web sites to package those data using the open source tools available from http://www.flaterco.com/xtide/files.html and distribute those data in TCD files.  Besides making the data traceable to authority with more direct quality control, this would eliminate the legal trap of determining whether reprocessing and redistributing those data for use with XTide is permitted by the terms of use:  such reprocessing and redistribution would be unnecessary if the data were already in a form that could be directly downloaded and utilized by end-users.

Failing that, the Sisyphean web-scraping job should be divided among N contributors working autonomously.  Any qualified volunteers out there who wish to maintain a harmonics database for their own countries may begin by clarifying the terms of use for the relevant data with their local tide authorities and then download the tools and go do it.  Following is a list of the data sources that have been brought to my attention.  If you succeed, let me know and I'll link to your download site. [**]

• Canada (I only processed the active stations; there are hundreds of old, inactive ones)
• Italy
• Norway
• Spain
• U.K.  (I processed these data once but would gladly hand off the maintenance task to someone local.)
• Miscellaneous
  • University of Hawai`i Sea Level Center
  • National Ocean Service (some locations outside of U.S. jurisdiction have historically been included along with the U.S. data)

* The German data were supplied as a rather short time series of water level observations from which I derived harmonic constants.  The agreement with official predictions was never that hot.  Further investigation by Jan Deepen in 2009 revealed significant, uncorrectable differences with authoritative predictions, so the German data were retired from the database at the end of 2009.

** Since a large part of the issue here is legal ambiguities, it would be preferable for you to retain control over your files on a server in your own country, where any copyright disputes might be resolved in a civilized fashion.

xtide

• SVG format support (Jack Greenbaum).
• Renovate to use the latest fad widget toolkit, use scalable graphics everywhere, play nice with modern desktop environments.
• Unbundle Dstr?

tcd-utils

• Bug:  build_tide_db loops forever if no <document> tag is found, or if the first line of the XML file is blank.  (Not a real parser.)

libtcd

• Things not to do unless/until a major revision is needed.
  • Get rid of internal state; fix inability to have more than one database open
  • Remove gratuitous complexity in the encoding of speeds, equilibrium arguments, and node factors; make speeds 32-bit unsigned with no offset
  • Possibly permit node factors to be specified for intervals shorter than 1 year and/or support apps that use libcongen to generate the node factors themselves

Harmbase2

• Clean up the code.

Suggested XTide features and other etceteras

The following features have been suggested or thought about but did not make the cut.  They could be added in future revisions if there were sufficient demand.

• There ought to be a way to specify relative dates and times in the –b and –e fields.  Need an applicable standard; ISO 8601 doesn't support it.  Simon Bridger requested –b and –e to recognize relative specifications like yesterday, today, tomorrow, mon–sun, sat+7 for next Saturday, "tomorrow 23:59" or "sat 00:00".  Others want to specify a small relative offset from the current time for –b to change the position of the + mark in graphs.
• Factor out a standalone component library as a favor to developers who want to reuse XTide's engine but reinvent its wheels.
• Generate maps for xttpd navigation and general illustration, include in LaTeX form.
• Graeme Rae suggested a line for the current time in text listings, like "2001-03-19 11:50 AM PST 0.10 feet Falling."  This is doable now that the tide events code has been refactored, but it's not clear what the settings and behaviors should be to handle the "now" event consistently across all modes.  Graphs and clocks have their own ways of showing the "now," and it would be inappropriate to include "now" in a calendar.
• Simon Bridger requested support for multiple –b and –e specifications.
• Tim Cera notes that Options might be more user-friendly if it was split into separate menus for "Create new windows" versus "Change settings."
• Add a real color chooser in control panel.  This would come free with Qt or similar if we ever migrated from Athena widgets.
• Constituent inference was patched in via libtcd and maybe could have been integrated better.  In theory, you might want to control it on a station-by-station basis like preferred units.  But, maybe it's a non-issue.  Nobody has complained yet.
• Generate node factors and equilibrium arguments more than once a year.  Most tide prediction software does it monthly or at least does it for the middle of your prediction interval.  But the legacy of SP98 is to do it yearly.
• Add support for Doodson style tide prediction as used by Foreman's IOS package.  At this time there seems to be little benefit to be gained by doing this:
  1. No new Doodson data appear to be forthcoming.
  2. Most Doodson constituents are approximated fairly well by Congen now.
  3. The ones that aren't approximated well are those that are drastically affected by latitude.  To support latitude-dependent constituents, node factors and equilibrium arguments would have to be generated internally to XTide, which would be a significant architectural change.
  4. Casement opined that the latitude-dependent method is bogus anyway because tides are generated some place in mid-ocean with a different latitude.
  5. If you want IOS, you can find it at http://www.pac.dfo-mpo.gc.ca/sci/osap/projects/tidpack/tidpack_e.htm.
• Simon Bridger requested a "weekend warrior" option that lists today, tomorrow, Saturday, Sunday, and next Saturday.
• Jef Poskanzer long ago asked for a global plot of tide levels to show how the tides move around.  This could be done by color-coding the dots on the globe, but it would (1) require a true-color display and (2) be too slow to be the default behavior of the globe.  Hans Bot has seconded this request.
• David Mendez suggests plotting the derivative of the tide as well as the tide—this is useful for predicting swells.
• Diane Grant wants to be able to execute a query like 'find all days in this year having a flood greater than 3.0 between 8 AM and 9 AM.'
• xttpd: Change mapping of locations to URLs so that links won't break when harmonics files are changed.  Thought about this but could not find a better solution than putting the entire location name in the URL, which is already semi-supported by the exact query feature.

← Previous Contents