Talk:Isotope

Maveric149 is planning on doing a massive enhancement to Wikipedia, putting in data for all 109 or so of the chemical elements, and I can hardly let him have all the fun, so I've started tinkering around with a related project. Over on http://www2.bnl.gov/CoN/nuchart1 are charts of all the known isotopes of every element. I've decided to add them into Wikipedia. I'm putting the first one below as a "working" copy, where I'll experiment until it's right. What does everyone think of it? It took me about half an hour to do, but I wouldn't want to do more than two or three of these in a day otherwise the boredom will become terminal. :) Bryan Derksen

Oh, one question for those who might know: on the original tables over at bnl, what is the significance of the half-colored isotopes? It will be tricky to duplicate the half-coloring in HTML, but if I know what it means I might be able to come up with some other method of indicating it. Other suggestions for improving the layout of the table are also welcome.

Wow! The table looks great! This only took you a half an hour? As far as I can tell it looks accurate. However, I am concerned that the table may be too big for low res screens - perhaps have the isotope number as a -1 font to save some space: ²⁵Si vs. ²⁵Si. Which isn't that much different for just this example, but multiplied by 12-20 times it would. I also think that the slightly smaller font looks better too (more like a superscript in a modern wordprocessor). Your color scheme seems to be very logical and a great improvement over the one on the BNL.gov site (I might have some suggestions later, but so far I can't think of any). I'm at work right now, but I'll pore over the table more this weekend to give you more input. Oh, and when the time comes, you are more than welcome to help me populate the new tables and headings with info in the elements articles - I wouldn't want to have all the fun just for myself. ;) --maveric149

I normally have an aversion to the <font> tag, but in this case screen space is at a premium so I'll give it a shot. :) As for the color scheme, my main concern is that the blue and indigo colors will blend into the color of linked text; see the deuterium and tritium links for example. When I get home I'm thinking of lightening the shades of all the colors, like was done with the background colors on the periodic table. I was originally going to be really geeky and reverse the spectrum of colors from the current version; longer wavelength of color would indicate longer half-life. But I think a purple-blue diagonal stripe is better looking than a bright red one, and "red/yellow/orange" = "highly radioactive" is a good way to go as well.

It only took half an hour thanks to Mozilla's HTML composer. Very good for roughing out initial table layout, though it needed a lot of editing to make the final code look nice. :) Bryan Derksen

I think the fields with two different colors indicate that the isotope can do two different decays, with different half-lifes. One could use a "subtable" (if the script is up to it;), or a small PNG as background (does that work?). As the other tables will go more to the upper right, leading table cells should probably be done with <td colspan=x>. --Magnus Manske

Oy, this little detail will be tricky. Why did nuclear physics have to be so complicated? :) I'll think about this some more and see what I can come up with, perhaps there will be some simple CSS trick I can use or something. As for the table's cell structure, colspan is a good idea, but on the other hand leaving all of the individual cells in there will make it easier to edit in the future. I still haven't decided on whether I should make the axes more "regular," for example, instead of scattering headers diagonally throughout like the original table did. I'll try both when I get home and have a better text editor than Wordpad available. Bryan Derksen

I thougth the scattering of the headers was quite odd at first. However, the way it is now makes it much easier to match up a particular row to an atomic number, or a column to a weight. I'd say keep the scattering. -- user:ansible

²⁶Al is both >10,000 years and unstable in the original table at BNL. Here's one attempt to convey this in HTML; colored borders. What do you think? Bryan Derksen

I was originally going to be really geeky and reverse the spectrum of colors from the current version; longer wavelength of color would indicate longer half-life.

Whoa! That's cool! I think you were on to something with that thought - Think about it: the most unstable stuff is the "hottest", right? As it is that stuff is white - and what is hotter than white hot? From there you go to progressively cooler colors for all the radioactive isotopes; violet, blue, green, yellow, orange, and then red (all paler than they are now). Then for the stable stuff, maybee have something like light gray. That way the colors are totally logical and easy to follow. The emphesis of color here should lean toward logic and be explanatory - this is a more geeky subject than the regular periodic table. Just my 2 cents. --maveric149

Alright, I'll reverse the spectrum when I put in the lighter versions of the colors. I'm thinking that I'll put that off until after I've finished all of the tables, however; I've been using Mozilla composer default colors, and it's really easy to just control-click a whole bunch of cells and select the color for them. It will be easy to do a global search-and-replace later. I'm already about 30% done the tables, so I'll just finish them all off in the next few days and then recolor. Bryan Derksen

Oh, for some reason the table below doesn't display well in Konqueror 2.2.1 - some of the superscripts are on top of the symbols, and then some are just superscripts (see example here (http://www.wikipedia.com/upload/maveric149-temp.png)). Looks just fine in Mozilla, IE, Galeon, and Navigator 4.7 though. So it is probably just an issue with KHTML - which should be fixed with the release of KDE 3.0 in a couple of months (hopefully fixing this, and giving konqi the ability to display -1 fonts -- every other non-text based browser I know of can). This might even be fixed in the 2.2.2 version - which I have been too lazy to compile. Anyway, just a heads up in case anybody complains. --maveric149

I note that virtually every two-letter element has the superscript "on top" and all of the single-letter ones are fine; the only exception is ⁴³Ti, which is quite thin and located in a column with a wide header. I think this is simply Konqueror's way of "squashing" the cells more. Bryan Derksen

Now, one other matter for discussion: where the heck am I going to put these things? One table per article, all in one big lump, or what? :) Bryan Derksen

---

I think there's no way to compress the tables as presented so that they'll be narrow enough to fit onto most browsers; some of the later elements have 20 or 30 isotopes listed each, so those tables are guaranteed to be at least that wide. So how about flipping the whole thing around, so that going from right to left increases proton number and going up increases neutron number? That way, each element gets its own column, and the table is only arbitrarily tall rather than arbitrarily wide. It won't be as easy to add new isotopes to it this way, but such is life. What does everyone think? I've put a transposed version of the second table below. Bryan Derksen

Very efficient use of space! You might still want to check out the uber page at bnl which shows the structure of all known isotopes for all the elements. It is here (http://www2.bnl.gov/CoN/). Other examples, http://atom.kaeri.re.kr/, http://t2.lanl.gov/data/map. However, this chemlab table (http://chemlab.pc.maricopa.edu/periodic/isotopes) is a good example of what an entire HTML table would look like stitched together. This table is really huge and shows every single isotope for each element. Although it's orientation is the opposite of bnl/yours...

The chemlab table is organized so that it starts at atomic number 1 (hydrogen) in the upper-left hand corner and then trends diagonally down toward the lower-right hand corner of the screen. The chemlab table starts at the lowest atomic number then as it trends down it is displaying increasingly large atomic numbers and nucleotide numbers. At first, just the opposite made sense to me: Atomic number 1 should be in the lower left, and atomic number 111 should be on the upper right (as it is with the bnl table (http://www2.bnl.gov/CoN/) and with your new format.

Unfortunately, if you look at the regular form of the periodic table you will see that it too starts with low atomic mass/numbers at the top-left and increasing atomic mass/numbers as you go down and across the table. In addition, when was the last time you clicked onto a webpage and ended up at the bottom of the page and had to read up? English just doesn't work this way. So, if you are really crazy, you might want to play with this. The end result (if stitched together) would trend in the same way as the chemlab table (http://chemlab.pc.maricopa.edu/periodic/isotopes) but have your clever column/rows switch.

But this still leaves us with the spacing issue.... Which I am glad to say has been solved by your novel solution of reversing the columns with the rows (which I hope wasn't difficult). It is kinda similar in concept to the Beryllium table -- you can add arbitrarily large amounts of info since the table grows in height, not width. This is a really good thing for scalability: For example, looking at the chemlab HTML table (http://chemlab.pc.maricopa.edu/periodic/isotopes), I see that Xenon has has about about 40 isotopes. This would be totally impossible to show as a single row displayable on anything but the largest monitors with the highest resolution without a horizontal scroll (on my home 14inch monitor with 1024x768 resolution, only about 20 of isotopes could be displayed at once). On the other hand, not all 40 would be displayable vertically either. This, of course, isn't as big of an issue since a web page naturally flows up and down in the vertical and this is the way English speakers read.

Another thing I like about your row/column switch, is that it now will be possible to have a single column in an isotope article aligned right with all the isotopes for that element in the column. Under your current example of the column/row switch, the largest number will be displayed at the top/first part of the page. However, if flip the example 180 degrees off the screen, then the opposite will be true: the isotopes will be listed with the lowest numbered one first and will trend down to the highest numbered ones (just as it now is with the isotopic part of the Beryllium properties table). I also think that having the bold symbols of each element on top will serve as a "backbone" that will help to visually organize the table (this is something that should be done regardless of your choice of going with the BNL or the chemlab trend). Hope this helps! --maveric149

Wow, lots of comments there. Basically, if I have this right, you're saing that the elements should be arranged with proton number increasing from left to right, and neutron number increasing from top to bottom? Yes, that makes good sense, and will actually save me some extra work when I do the table-reversal thing (I downloaded a new piece of software to do it with, Namo WebEditor 5 trial version, and even so there's a fair amount of dragging rows and columns into new positions manually). I've been continuing to create the tables initially in the same orientation as the bnl tables, since it makes error-checking them and coloring them so much easier. I'm up to Technetium now. :)

Once these tables are finished, I think it would be reasonable to paste them together into standardized browser-window-width chunks as well as one monster-huge table for a separate article (like the huge alternate periodic table). The browser-window-sized chunks could go at the bottom of the main Isotopes article without much problem; scrolling up and down is quick and easy. Bryan Derksen

---

Per your edit comment question: Yes I think width is starting to become an issue again. It's those damn elements with more than 100 nucleons that are causing it. Perhaps have 10 or 15 columns instead of 20... BTW would it be possible to set the width of the first few sub-tables (which naturally take up less real estate due to lower nucleon counts and one letter symbols) to match that of the heavy weights that have triple digit nucleon counts and three-letter symbols? That way, everything would line up nicely and would be easier to follow. --maveric149

I'll switch to 15 each. The current table for elements 41-60 is about as large as a 20-element table is likely to get (all columns now have triple digit nucleon numbers and most of the elements have double-letter symbols), and on my monitor it looks like shaving 5 elements off of the end will make it fit just right. I'll be able to get the thinner tables to match up by setting every table's width="100%", which should stretch them out to fill the available area. I'll see if I can make the columns equal width, too. Bryan Derksen

---

Alright, here finally are the 15-column tables. I haven't done the colspan stuff for the empty cells, but I'm not going to until I know these fit right this time. :) Bryan Derksen

working area

Since 90% of this talk page consisted of table code that's now obselete, I've deleted that stuff. See the completed tables linked to from Isotope for future discussion.