Communities

Writing
Writing
Codidact Meta
Codidact Meta
The Great Outdoors
The Great Outdoors
Photography & Video
Photography & Video
Scientific Speculation
Scientific Speculation
Cooking
Cooking
Electrical Engineering
Electrical Engineering
Judaism
Judaism
Languages & Linguistics
Languages & Linguistics
Software Development
Software Development
Mathematics
Mathematics
Christianity
Christianity
Code Golf
Code Golf
Music
Music
Physics
Physics
Linux Systems
Linux Systems
Power Users
Power Users
Tabletop RPGs
Tabletop RPGs
Community Proposals
Community Proposals
tag:snake search within a tag
answers:0 unanswered questions
user:xxxx search by author id
score:0.5 posts with 0.5+ score
"snake oil" exact phrase
votes:4 posts with 4+ votes
created:<1w created < 1 week ago
post_type:xxxx type of post
Search help
Notifications
Mark all as read See all your notifications »
Meta

Comments on How to request a mass-change in capitalization for tags?

Post

How to request a mass-change in capitalization for tags?

+2
−1

In my effort to clean-up and organize tags on EE.CD I arrived at a point where there are a lot of old tags with wrong capitalization.

How should I request the staff to change them?

Should I post here on meta a list (maybe as an edit in this post) with something like this:

old --> NEW

for example:

led --> LED
Wireless --> wireless
triac --> TRIAC
...
etc.

Or should I mark the tags with some text in its guidance text or WIKI text, so that the devs could handle them using a script that searches the DB entries for some canned text like (for example):

CHANGEINTO:"LED"
CHANGEINTO:"wireless"
CHANGEINTO:"TRIAC"

And if this latter approach is good, how do I ping the devs to tell them I'm done and they could implement the change? Do I post another question? Do I flag this question?

Thanks for the help!


EDIT

The tags to be renamed (old vs. new names are listed below). The reason why I ask for the change is to make them complying with the tag guidelines I proposed and the community seemed to agree with (post here).

BTW, in the list there are also tags that should be deleted because violate those guidelines. Some show a non-zero question count but most are deleted questions. I marked them with "DELETE" below.

  • ac --> AC
  • ac-circuit --> AC-circuit
  • ac-power --> AC-power
  • adc --> ADC
  • Analog --> DELETE
  • Antenna --> antenna
  • arm --> ARM
  • bode-plot --> Bode-plot
  • can-bus --> CAN-bus
  • clk --> DELETE
  • coil --> DELETE
  • tvs --> TVS
  • diac --> DIAC
  • triac --> TRIAC
  • bjt --> BJT
  • jfet --> JFET
  • dac --> DAC
  • dc --> DC
  • dc-dc --> DELETE
  • dc-generator --> DC-generator
  • dma --> DMA
  • esd --> ESD
  • darlington --> Darlington
  • emi --> EMI
  • esd-protection --> ESD-protection
  • esl --> ESL
  • esr --> ESR
  • Pi-filter --> pi-filter
  • rlc-filter --> RLC-filter
  • ground --> DELETE
  • high-speed --> DELETE
  • Intrinsic-Safety --> intrinsic-safety
  • ipc --> IPC
  • library --> DELETE
  • losses --> DELETE
  • master --> DELETE
  • math --> DELETE
  • measurements --> measurement (merge with existing tag; "measurement" survives)
  • Mixer --> mixer
  • mosi --> DELETE
  • pmsm --> PMSM
  • no --> DELETE
  • oamp --> DELETE
  • op-amp --> opamp (merge with existing tag; "opamp" survives)
  • ota --> OTA
  • pain --> DELETE
  • plc --> PLC
  • PMSM-motor --> DELETE
  • pole --> DELETE
  • smps --> SMPS
  • ppg --> PPG
  • c --> C
  • vhdl --> VHDL
  • PTC --> DELETE
  • ram --> RAM
  • rf-switch --> RF-switch
  • sar --> SAR
  • scattering-coefficient --> scattering-coefficients
  • Shoot-through --> shoot-through
  • smd --> SMT
  • software --> DELETE
  • solid-state --> DELETE
  • solids --> DELETE
  • spi --> SPI
  • spice --> SPICE
  • thermal --> DELETE
  • usb --> USB
  • uvlo --> UVLO
  • vfd --> VFD
  • wifi --> WiFi
  • Wireless --> wireless
  • zero --> DELETE

EDIT

In response to Lundin's answer.

Your objections are in part valid but I think some of them are based on false premises. Moreover I think that maybe you miss the implications of having a tag hierarchy facility and tag synonyms (more about this below).

SMD is, I believe, a more common term than SMT. Both are commonly used. These should be synonyms, as should surface-mount-device and surface-mount-technology.

MOSI is a common, de facto standard name for a SPI signal. There's no reason to delete it.

"op-amp" should probably be made synonymous for operational-amplifier.

Please, bear in mind that my work on tags is still ongoing and some of the deletion I suggested are just an intermediate step.

For example, just to address some of your criticism: mosi was to be deleted to be next re-added as a "synonym" for the SPI tag (as a standalone tag is absolutely too specific, and outside of SPI context has no meaning).

The same is for SMT: I was waiting for the devs to make those modifications, so that I could re-add SMD to SMT as a synonym (technically SMT is a more general term, but having SMD as a child tag of SMT is too specific, IMO).

opamp is already a synonym for operational-amplifier (actually opamp is the main tag because it was already there before I added operationa-amplifier). op-amp will be added as synonym as soon as the devs apply the changes I requested.

If you haven't already, please check the hierarchy button in the tags page and see the effect of my (still ongoing) work on building a tag structure. That would be impossible on SE and would be impossible if tags were treated simply as keywords.

Whereas I don't think that "AC-circuit" and "AC-power" are necessarily well-established terms. The first can mean anything, the latter should perhaps be tagged VAC.

Please, keep in mind that the work is ongoing, so I tackled the most egregious issues first. You are right that there are still tags that are debatable like AC-circuit, but you can tell that it wasn't something I created because it doesn't have a guidance text (I did create a few tags without guidance text, but they are quite specific and they were put in a hierarchy).

Tags like AC-circuit are used for a lot of questions and I will need the community guidance to disentangle them, so I postponed those until I finished the "grunt work".

I will post on meta about the destiny of such tags to have feedback once the most of the less controversial issues are solved.

First of all, please note that there is no criteria stating that a tag must be able to "stand alone" - sometimes a topic is filtered out by a combination of tags. For example the tag analog may be an "adjective tag" but it could be meaningful in combination with various other tags, to narrow down the scope to a specific term.

Sorry, I disagree in this. I went on with some modifications because it seemed the community had reached some consensus. I posted a proposal some time ago which seemed to be well received. I also stated clearly that my premise was that for me tags were not a "keyword" facility but something that should help form hierarchies of related concepts (i.e. "structure").

That's a whole different approach from your "tags are keywords" POV. I understand that the choice is debatable, that's why I asked on meta, and no one objected to what I proposed (even Olin and Monica Cellio agreed explicitly commenting on that post).

Please, bear in mind that your vision of how we should use tags, while a perfectly acceptable POV (not mine, though), is not applicable effectively to our facility because we have a hard limit to 5 tags. A complex question involving many EE areas could not be tagged effectively, because if tags were simply keywords the post would need to be tagged with dozens of tags.

And those are questions that could also be very good questions (well researched and with lots of info)!

Moreover, tags as keywords aren't really adding any value IMO, because if you wanted to search for anything that puts together "analog" and "ground" it would simply suffice to use the search bar with those two terms. Tags as keywords would simply duplicate this (and possibly in a worse way).

On the contrary, having a complex, curated tags structure will allow something that is not possible with keywords: searching for logical relationships between questions.

Compare this tag sequence:

analog-comparator, digital-ground, safety-ground, DC-motor

to this one (assuming you could use more than 5 tags):

analog, digital, ground, safety, DC, motor, comparator

The first one gives quite a precise description of the topic of the questions. However, the second could be applied to a huge number of unrelated posts.

The main criteria for tags: they must be related to electrical engineering and they must correspond to commonly used terms in electrical engineering, physics or applied electronics. It need not necessarily be the best or most technically correct term.

I largely agree, but that's why synonyms are so useful. For example I explicitly added scope as a synonym to oscilloscope for this same reason.

History
Why does this post require moderator attention?
You might want to add some details to your flag.
Why should this post be closed?

2 comment threads

Renames are done (see the edit in my answer). (1 comment)
I see your edit with the list and also that there's some disagreement. Please reply to this comment ... (20 comments)
I see your edit with the list and also that there's some disagreement. Please reply to this comment ...
Monica Cellio‭ wrote 9 months ago

I see your edit with the list and also that there's some disagreement. Please reply to this comment when issues are resolved; I want to help with your cleanup efforts and don't want to act prematurely. Thanks.

Lorenzo Donati‭ wrote 9 months ago

Monica Cellio‭ Frankly, I don't know how to proceed. The only explicit disagreement is from high-rep user Lundin‭, to the objections of whom I replied with my edit, at least mostly. They didn't reply in turn, yet.

I identify essentially these major objections: (A) that I shouldn't have acted without consensus with such a major change; (B) that my modifications were disruptive of the previous situation, especially because note that there is no criteria stating that a tag must be able to "stand alone" - sometimes a topic is filtered out by a combination of tags (this approach is a "keyword POV"), (C) that my continuous changes created a lot of "post bumping" and that was disruptive as well; (D) and that I should have also fixed other issues with the bumped posts.

Lorenzo Donati‭ wrote 9 months ago

Monica Cellio‭

What I could add now is that objection (A) is not warranted: I did ask the community about (1) tag guidelines (proposed by me) and (2) interpretation of the tags feature (at the end of my post). I received no objection at all, even from Lundin ("I agree with everything said."). So I have inferred that (1) those guidelines are OK for the community, (2) the tags are not a list of keywords, but they must help build a structure (i.e. a network of semantic relationships) (rebuke of part of objection B).

Lorenzo Donati‭ wrote 9 months ago

Monica Cellio‭

On the basis of that inference I put some effort into changing the status quo (objection B), a situation that was incredibly depressing: before I began my work the tags page had not even one guideline text, WIKI, synonym or parent. At all. What I got from that was that (non exclusively): no one use tags for searching (so they didn't see those major issues), or no one gives a rat's ass about tagging in a useful way, no one even cares to correct or flag blatant issues like the following tags big-honking-spark, cock or shit. Well, I think that disrupting that situation was good.

As for objection (C), that's warranted for sure, but since we don't have a mechanism to voluntarily avoid post bumps, my answer is "you can't make an omelette without breaking the eggs". I thought some bumping was an unavoidable "collateral damage" to solve what I saw as a big issue.

Lorenzo Donati‭ wrote 9 months ago · edited 9 months ago

Monica Cellio‭

As for the objection (D), sorry, definitely no: I'm not going to spend time correcting issues with any single post I retag. Some of those posts shouldn't even exist. BTW, I don't think there's a rule like on SE that mandates an edit being necessarily complete (maybe I missed that somewhere).

Everything I did was public and also seemed supported by the mods. No one has "screamed at me" to stop and that is the only post that openly objects.

Frankly doing janitorial work on something no one seems to have given some love to until now is not particularly enthusing. If the process also becomes clunky, then I can stop and devote the time I spare to write my posts.

What should I do?

Monica Cellio‭ wrote 9 months ago

I see the frustration in your comments, and I understand -- we all want things to be better, you're doing a lot of good work, and there was prior discussion. I think we're seeing a little scope drift -- totally natural given the scale of the work! -- and maybe we need to divide and conquer. The earlier question, and the title of this post, is about capitalization, so does it make sense to separate the list of proposed changes into capitalization and everything else? We can then knock out the capitalization ones (they sound uncontroversial), and figure out what other discussions need to happen for the other ones. Would that make sense?

Lundin‭ wrote 9 months ago · edited 9 months ago

I think the main issue here is that there's no clear consensus regarding tags that aren't meaningful stand-alone but otherwise clearly on-topic. As I already tried to explain here https://electrical.codidact.com/posts/289217/289280#answer-289280, such tags might give the post a meaning in combination with other tags. It isn't obvious to me how deal with the situation. coil and voltage are examples of such vague tags and spontaneously one might think to address that by creating a new tag coil-voltage if that's what a post is about. But I don't think that is viable either, or soon we will have a tag for every row of electrical characteristics in every datasheet for any component. coil-voltage is by no means any more an established term... but it makes perfect sense if used in combination with relay. Are we to create yet another tag relay-coil-voltage then - see where this is going?

Lundin‭ wrote 9 months ago

I think we have to live with on-topic tags that hold no meaning of their own and this is how SE always worked too.

Lundin‭ wrote 9 months ago

But also we should not make it harder to change tags than to invent them. If we allow everyone and their mother to create tags, which is usually the case, I don't think we can regard tags as something "holy" as on SE but let trusted users/moderators change them without too much meta debating. I think Lorenzo is a trusted user and probably have a better clue of what makes a good tag than the majority of the people who invented all of those, so I have no problem with them cleaning up tags on case by case basis. We only need to have a unison policy about the mentioned "vague" tags, because if we get rid of voltage and the like they will surface again within a week.

Elleanor Lopez‭ wrote 9 months ago

Lorenzo Donati‭ Is there some way you can edit these tags without turning the EE forum upside down? Each tag edit floats old questions to the top of the feed.

Lorenzo Donati‭ wrote 9 months ago

Monica Cellio‭ Thanks for your words, I really appreciate them. As for your proposal, that's OK for me. Do perform just the capitalization changes, they are needed anyway. Although it's the other tags deletion/merging that would allow me continue reorganizing the existing tags (some of those would have been blacklisted, some other would have been re-added as synonims).

Lorenzo Donati‭ wrote 9 months ago

Monica Cellio‭ After I read your comment I realized I did do a mistake: I got the title of the post wrong (sorry for that). As you note, I said "capitalization change", whereas I also proposed deletion.

Thinking about it, I can recollect the genesis of that title. I just wanted to ask for capitalization changes, and I started writing the post on an external editor (I do this when I need to cherry pick info from various places), then when exploring the tags list for the nth time I noticed that there were some tags that needed deletion as well.

I thought it was good if I asked for all the changes in just one post. The problem is that I didn't update the title and the text to also explicitly mention deletion. I was too busy building the list of changes and probably I had in mind to do that at one point, but was carried away and didn't re-check to see I've actually implemented the change. Having built the original question in pieces at different times didn't help. Oh well!

Lorenzo Donati‭ wrote 9 months ago

Elleanor Lopez‭ Sadly it's not possible for a user making an edit to avoid that "bump". There are discussions in CD.META to implement a checkbox to do that, but it's not there. I think only devs can change things without bumping the question. A workaround is to sort the questions by age, so you see them in the reverse order they were posted, but of course you will miss edits. I know this is a major annoyance, but there is no other way to perform massive edits. Any major clean-up activity is highly disruptive, unfortunately.

Lorenzo Donati‭ wrote 9 months ago

Lundin‭ First of all, thanks for your feedback.

"I think we have to live with on-topic tags that hold no meaning [...] But also we should not make it harder to change tags than to invent them. If we allow everyone and their mother to create tags, which is usually the case, I don't think we can regard tags as something "holy" as on SE but let trusted users/moderators change them without too much meta debating." I totally agree with this (BTW, my "tier-2" tag privilege proposal goes in that direction).

Lorenzo Donati‭ wrote 9 months ago · edited 9 months ago

Lundin‭ "[...] such tags might give the post a meaning in combination with other tags." Here I strongly disagree. I see your POV of as "keywords" instead of as "semantic links" (my view). But there is a big problem when you say "in combination with other tags". In the current tag system there is no notion of "combination". You make the example of coil and voltage that when used together would convey the same meaning of coil-voltage. But this is misleading, IMO; it's just an artifact of the GUI; there is no actual combination.

Lorenzo Donati‭ wrote 9 months ago

Lundin‭ What if you would want to convey the meaning of coil-voltage and capacitor-current in the same post? You would use the tags coil, voltage, capacitor, current. But there is no mechanism that guarantees the tags will be always displayed that way (Maybe devs will decide tags will be displayed in alphabetical order in the future).

Not only retagging could change the order, but people have no clue that that association is the one you want to convey. Someone could reasonably think the post is about coil current too.

Keywords systems are intrinsically ambiguous, unless there is some codified mechanism to express relationships between words. That's the same problem a computer language parser has, and the solution there is a formal grammar that can tell the compiler that, using a C example, const int * variable is not the same type as int * const variable (just a silly example to make my point more concrete).

Lorenzo Donati‭ wrote 9 months ago · edited 9 months ago

Lundin‭ But see, int const * is the same type as const int * (due to a quirk of the C grammar), so you have a blatant visual ambiguity, which is resolved behind the scenes by a semantic rule. A person not knowing the grammar faced with the three alternatives int const *, const int * and int * const could well think they express all the same concept, or that they express three different concept. Instead the reality is that there are just two concepts!

These are the pitfalls of just juxtaposing words and thinking their juxtaposition intrinsically conveys semantic meaning. And with human languages that is much worse, of course. That's why I'm opposed to a "keyword" approach. Moreover, as I said in an edit, you could just use the search box to search for keywords anyway, so tags would be just a sup-par duplicated feature.

Lorenzo Donati‭ wrote 9 months ago · edited 9 months ago

Lundin‭ I can see your concern about having an explosion of tags like relay-coil-voltages, but that's why tag curation is important to identify "too-specific" tags. In my "semantic tag" view a tag should have the right to exist if it is specific enough to identify a reasonably defined area of expertise, but not so specific to be of interest of just "one person" (or to be used to mark potentially too few answers).

Yes, there are human judgement calls to be made to keep the system healthy, so that's another area where active curation/moderation would be needed. However the benefits would be quite substantial for the learner: they could be led to one question by chance, but clicking on a tag they could discover a whole lot of related questions that would help them in their research/study. And they would know those questions are really related because a human has (ideally) checked that they were so.

Lundin‭ wrote 9 months ago

Lorenzo Donati‭ But if we look at more specific tags, you also propose to delete PTC, which stands for positive temperature coefficient. This term is (almost?) exclusively used in the context of thermistors, at least as far as I know. Therefore PTC is a clarification of a question tagged thermistor in case it matters. Or NTC could be used. Or neither could be used, in case the question is about thermistors in general. Yes this could be solved by creating 3 different tags PTC-thermistor, NTC-thermistor and thermistor but the total amount is still 3 and nothing was gained from getting rid of PTC in the first place. Furthermore, engineering jargon often just calls them PTC/NTC - if you look at Li/Ion charger IC for example they typically have a pin just labelled NTC and the reader is implicitly assumed to understand what it means.

Lundin‭ wrote 9 months ago

Lorenzo Donati‭ Ok so I tried to compile a draft for some sort of general policy here: Tag creation/deletion crieria. And yeah this might mean that tags such as voltage are free to delete. Any feedback is appreciated.