Ok, then I will put my anal personality (which would never agree to such a thing) aside and say I'm cool with this. On Mon, Dec 2, 2013 at 8:43 PM, Matthew Turk wrote:

Hi Britton,

This would be in the units refactor of yt-3.0, and would *only* apply to parsecs. This would be the exclusive special case of m[some unit] meaning Mega.

-Matt

On Mon, Dec 2, 2013 at 3:41 PM, Britton Smith wrote:

...

Hi Matt,

Where exactly is this? Is this in yt-3.0, and if so, would this apply to all uses of m as milli? Is that is so, then my concern is for when yt becomes more used by other domains that actually do have a valid use for milli.

Britton

On Mon, Dec 2, 2013 at 8:24 PM, Matthew Turk wrote:

...

Hi all,

Nathan and I are working on some unit things. One that came up is that it looked like the conversion to mpc was wrong.

Turns out, it's not. But what it's set up as is "mpc" for "milliparsecs" and "Mpc" for "Megaparsecs." We'd like to get rid of any usage of milliparsecs, since I think the number of people who want to use that (globally) are probably between 0 and 0, and just make Mpc and mpc the same thing.

If anybody objects, write back. Silence will be taken as tacit

agreement!

...

-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

-0 I'd rather we all just use Mpc, but I'll go along with it. On Mon, Dec 2, 2013 at 12:53 PM, Nathan Goldbaum wrote:

Here is the change:

https://bitbucket.org/ngoldbaum/yt/commits/2e6d536ffa1976e4699093c4bb3b3bc4a...

We'll definitely be leaving 'm' -> 'milli' alone. FWIW, we already do this by mapping 'mpc' -> 'Mpc' in plot labels.

On Mon, Dec 2, 2013 at 3:46 PM, Britton Smith wrote:

...

Ok, then I will put my anal personality (which would never agree to such a thing) aside and say I'm cool with this.

On Mon, Dec 2, 2013 at 8:43 PM, Matthew Turk wrote:

...

Hi Britton,

This would be in the units refactor of yt-3.0, and would *only* apply to parsecs. This would be the exclusive special case of m[some unit] meaning Mega.

-Matt

On Mon, Dec 2, 2013 at 3:41 PM, Britton Smith wrote:

...

Hi Matt,

Where exactly is this? Is this in yt-3.0, and if so, would this apply to all uses of m as milli? Is that is so, then my concern is for when yt becomes more used by other domains that actually do have a valid use for milli.

Britton

On Mon, Dec 2, 2013 at 8:24 PM, Matthew Turk wrote:

...

Hi all,

Nathan and I are working on some unit things. One that came up is that it looked like the conversion to mpc was wrong.

Turns out, it's not. But what it's set up as is "mpc" for "milliparsecs" and "Mpc" for "Megaparsecs." We'd like to get rid of any usage of milliparsecs, since I think the number of people who want to use that (globally) are probably between 0 and 0, and just make Mpc and mpc the same thing.

If anybody objects, write back. Silence will be taken as tacit

agreement!

...

-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

Well, count me as a not terribly passionate -1. I think the idea that coding around the habits of certain subsets of users of yt has gotten us into trouble in the past, and I'm wary of doing it here. While this is certainly not a huge deal, I personally prefer that we be stricter about units standards. However, I totally understand that milliparsecs is not used by anyone. And of course, parsecs are not SI to begin with, so I suppose from a PURELY logical standpoint, we aren't necessarily bound by their prefix conventions. j On Mon, Dec 2, 2013 at 4:22 PM, Sam Skillman wrote:

-0

I'd rather we all just use Mpc, but I'll go along with it.

On Mon, Dec 2, 2013 at 12:53 PM, Nathan Goldbaum wrote:

...

Here is the change:

https://bitbucket.org/ngoldbaum/yt/commits/2e6d536ffa1976e4699093c4bb3b3bc4a...

We'll definitely be leaving 'm' -> 'milli' alone. FWIW, we already do this by mapping 'mpc' -> 'Mpc' in plot labels.

On Mon, Dec 2, 2013 at 3:46 PM, Britton Smith wrote:

...

Ok, then I will put my anal personality (which would never agree to such a thing) aside and say I'm cool with this.

On Mon, Dec 2, 2013 at 8:43 PM, Matthew Turk wrote:

...

Hi Britton,

This would be in the units refactor of yt-3.0, and would *only* apply to parsecs. This would be the exclusive special case of m[some unit] meaning Mega.

-Matt

On Mon, Dec 2, 2013 at 3:41 PM, Britton Smith wrote:

...

Hi Matt,

Where exactly is this? Is this in yt-3.0, and if so, would this apply to all uses of m as milli? Is that is so, then my concern is for when yt becomes more used by other domains that actually do have a valid use for milli.

Britton

On Mon, Dec 2, 2013 at 8:24 PM, Matthew Turk wrote:

...

Hi all,

Nathan and I are working on some unit things. One that came up is that it looked like the conversion to mpc was wrong.

Turns out, it's not. But what it's set up as is "mpc" for "milliparsecs" and "Mpc" for "Megaparsecs." We'd like to get rid of any usage of milliparsecs, since I think the number of people who

want

...

to use that (globally) are probably between 0 and 0, and just make Mpc and mpc the same thing.

If anybody objects, write back. Silence will be taken as tacit agreement!

-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

I'm changing my vote to somewhere between -0 and -1. We will still be supporting many other implausible units, like centi and nanoparsecs, which makes this wrinkle seem even more out of place. I can only see two upsides to making this change: backward compatibility and difficulty/laziness with identifying and fixing all the instances of 'mpc' in the code and docs. Backward compatibility has already largely been thrown aside for 3.0. Other people's reservations have brought me around and I now see the second upside as not really worth it. On Tue, Dec 3, 2013 at 1:52 AM, Cameron Hummels wrote:

I'm +0 on having mpc map to Mpc. I think this is a good situation where we could have a note in the docs about the case-sensitivity of units in the gotchas section of units or fields or some such.

On Mon, Dec 2, 2013 at 2:32 PM, j s oishi wrote:

...

Well, count me as a not terribly passionate -1. I think the idea that coding around the habits of certain subsets of users of yt has gotten us into trouble in the past, and I'm wary of doing it here. While this is certainly not a huge deal, I personally prefer that we be stricter about units standards. However, I totally understand that milliparsecs is not used by anyone. And of course, parsecs are not SI to begin with, so I suppose from a PURELY logical standpoint, we aren't necessarily bound by their prefix conventions.

j

On Mon, Dec 2, 2013 at 4:22 PM, Sam Skillman wrote:

...

-0

I'd rather we all just use Mpc, but I'll go along with it.

On Mon, Dec 2, 2013 at 12:53 PM, Nathan Goldbaum wrote:

...

Here is the change:

https://bitbucket.org/ngoldbaum/yt/commits/2e6d536ffa1976e4699093c4bb3b3bc4a...

We'll definitely be leaving 'm' -> 'milli' alone. FWIW, we already do this by mapping 'mpc' -> 'Mpc' in plot labels.

On Mon, Dec 2, 2013 at 3:46 PM, Britton Smith wrote:

...

Ok, then I will put my anal personality (which would never agree to such a thing) aside and say I'm cool with this.

On Mon, Dec 2, 2013 at 8:43 PM, Matthew Turk wrote:

...

Hi Britton,

This would be in the units refactor of yt-3.0, and would *only* apply to parsecs. This would be the exclusive special case of m[some unit] meaning Mega.

-Matt

On Mon, Dec 2, 2013 at 3:41 PM, Britton Smith wrote: > Hi Matt, > > Where exactly is this? Is this in yt-3.0, and if so, would this apply to > all uses of m as milli? Is that is so, then my concern is for when yt > becomes more used by other domains that actually do have a valid use for > milli. > > Britton > > > On Mon, Dec 2, 2013 at 8:24 PM, Matthew Turk wrote: >> >> Hi all, >> >> Nathan and I are working on some unit things. One that came up is >> that it looked like the conversion to mpc was wrong. >> >> Turns out, it's not. But what it's set up as is "mpc" for >> "milliparsecs" and "Mpc" for "Megaparsecs." We'd like to get rid of >> any usage of milliparsecs, since I think the number of people who want >> to use that (globally) are probably between 0 and 0, and just make Mpc >> and mpc the same thing. >> >> If anybody objects, write back. Silence will be taken as tacit agreement! >> >> -Matt >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

Okay. Y'all convinced me. Let's revert and fix it to be consistent. On Dec 3, 2013 7:28 AM, "Kacper Kowalik" wrote:

On 12/02/2013 09:53 PM, Nathan Goldbaum wrote:

...

Here is the change:

https://bitbucket.org/ngoldbaum/yt/commits/2e6d536ffa1976e4699093c4bb3b3bc4a...

...

We'll definitely be leaving 'm' -> 'milli' alone. FWIW, we already do

this

...

by mapping 'mpc' -> 'Mpc' in plot labels.

That's precisely the reason it shouldn't be done this way. You're writing additional lines of code to implement workaround for invalid exception from general rules.

I'd give strong -1 to the idea and additionally s/mpc/Mpc/g through out the codebase. Of course bear in mind that this comes from a person that never did simulation larger than 20kpc :P Cheers, Kacper

...

On Mon, Dec 2, 2013 at 3:46 PM, Britton Smith

...

Ok, then I will put my anal personality (which would never agree to

such a

...

...

thing) aside and say I'm cool with this.

On Mon, Dec 2, 2013 at 8:43 PM, Matthew Turk

...

Hi Britton,

This would be in the units refactor of yt-3.0, and would *only* apply to parsecs. This would be the exclusive special case of m[some unit] meaning Mega.

-Matt

On Mon, Dec 2, 2013 at 3:41 PM, Britton Smith wrote:

...

Hi Matt,

Where exactly is this? Is this in yt-3.0, and if so, would this apply to all uses of m as milli? Is that is so, then my concern is for when yt becomes more used by other domains that actually do have a valid use for milli.

Britton

On Mon, Dec 2, 2013 at 8:24 PM, Matthew Turk wrote:

...

Hi all,

Nathan and I are working on some unit things. One that came up is that it looked like the conversion to mpc was wrong.

Turns out, it's not. But what it's set up as is "mpc" for "milliparsecs" and "Mpc" for "Megaparsecs." We'd like to get rid of any usage of milliparsecs, since I think the number of people who

want

...

to use that (globally) are probably between 0 and 0, and just make Mpc and mpc the same thing.

If anybody objects, write back. Silence will be taken as tacit agreement!

-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org