JSON Resume – a JSON-based open source standard for resumes

by krrishd • 12 years ago

490 210 12 years ago

onassar• 12 years ago

Awesome idea. I could imagine a marketplace for resume themes that is based on this. Would be easy to switch between different designs to see consistent details.

Well done.

blackaspen• 12 years ago

This. This is it.

I've actually written my resume in markup for a while just so that I could put it on a nice dynamically driven site. Sort of a fun weekend project (no, it's not hosted anywhere).

While I strongly doubt that this will ever be a standard, I think it's nifty, and potentially useful. It opens up easy, pretty resumes, which the world really needs (Note: I'm not being sarcastic. The number of mountains-of-text word document resumes I've gotten in the past few weeks is too high).

xienze8• 12 years ago

It opens up easy, pretty resumes, which the world really needs

People have been doing XML+XSLT for their resumes for a long time, yet it really hasn't taken off in a standard way for whatever reason. Not sure this being JSON-based will change that. I guess I'm wondering what this solution brings to the table that previous attempts didn't.

anko• 12 years ago

Isn't the look of your resume one of your competitive advantages as job-seeker?

Sorry I really don't understand the use-case. Having an API is good for styling (as you say), or doing some soft of programmatic transformation to data. I am having trouble understanding the need to do either for a resume.

Maybe it would help as metadata for a search engine, but in that case I think a plain text search would be just as effective?

I think json schema is a really nice thing to have and I'd like to see more people adopt it, but I'm having trouble coming up with the point for this use.

aroman• 12 years ago

Came here to say this. I feel the same way — even about the schema. Having a poorly-organized or thought out resume should be an immediate signal to employers. This eliminates that.

Of course, this is not to imply that having a really well-done resume is really worth much either.

etimberg• 12 years ago

It's already possible using LaTeX and Modern CV http://www.ctan.org/pkg/moderncv

mitchellh• 12 years ago

I looked over the website briefly, and I couldn't find an answer to this question so I'm going to ask here. I am not asking this tongue-in-cheek, these are serious questions:

* By "open source standard," what do they mean? I don't see any evidence they're on an RFC track, so this would be "standard" in what way?

* Why? I've never needed a resume in JSON format, but that is anecdotal. They don't really say why they're pursuing this on the site, except why they chose JSON.

Good luck, either way! Making something isn't easy, so I applaud any effort to do anything.

gkoberger• 12 years ago

Assuming this takes off and everyone uses it (you be the judge of those odds), I can see a standard format being really useful for both employers and job board websites.

Let's say you had a bank of resumes -- you could easily find all resumes that know Java and have had at least 2 jobs and an active GitHub repo, and automatically send out a request to their references.

Currently, you have to read an unformatted PDF and extract the info manually -- it's not the end of the world, but it gets tedious if you're dealing with lots of resumes.

rcfox• 12 years ago

you could easily find all resumes that know Java

Except that you're going to have people who put "J2EE" or "J2ME" or "JavaFX" or whatever else people like to specialize in instead of plain "Java".

hmsimha• 12 years ago

It would be really useful for the people making the resume also. Right now the current mainstream advice is to use Microsoft Word for creation and editing and then (typically) to submit as PDF. Both of those document standards are proprietary and opaque. If you two computers that don't have the same operating system, you're probably SOL for tweaking your resumes across computers as well.

thomasfromcdnjs25• 12 years ago

You are right, I am still researching how to manage a standard formally hence why the site wasn't meant to go on HN just yet. If you have any good advice or links to advice for writing and managing standards that would be fantastic.

For example, what is an "RFC track"? I was also hoping that Github Issues might be able to replace the mailing list style of managing standards, what do you think?

M2Ys4U• 12 years ago

I assume they're talking about the Iternet Standards Process[0] where documents are published as IETF RFCs.

[0] http://tools.ietf.org/html/bcp9

thomasfromcdnjs25• 12 years ago

Hey guys, we weren't actually ready for a lot of eyes to see this project. We were hoping to keep it small and get some feedback on important decisions about the project's trajectory.

The plan is to create a standard that developers worldwide are comfortable enough with to build upon.

To give the standard some traction, we have built tools to export to other formats such as pdf, txt and doc.

We also thought it would be cool and useful to have an NPM like system for resumes. So we built a CLI heavily inspired by NPM, to init and publish your resume to the jsonresume.org registry.

The themes right now are in bad shape, and we are looking for designers who would love to jump in during the early stages. We are thinking of building a theming version manager on top of NPM to take advantage of versions/distribution and also allow theme developers to implement their theming systems however they like (erb, mustache, md etc).

All the code is open source and available at -> http://github.com/jsonresume

Theme developers should be looking at the resumeToHTML repo -> http://github.com/jsonresume/resumeToHTML

People with comments about the standard semantics and structure are encouraged to post their ideas to Github issues. -> http://github.com/jsonresume/resume-schema We also want to formalize the standard a lot more over the coming weeks.

If the project can be pulled off professionally and driven by the community, we believe that working with resumes will be fun!

Edit: Here are some of the reasons why I personally think JSON resumes will improve HR.

(trying to figure out bullet points)

- Better searching and filtering than PDF/doc. Making recruiting a tad easier.

- Applying themes will be a lot easier.

- Applying for jobs will be easier. Instead of filling out complex forms over and over again services might start allowing you to auto-fill with your JSON resume.

- There are many services that ask for your career history such as LinkedIn, Angel.co etc. These could be autopopulated or synced with your master copy.

- You can update your resume programatically so Github projects can be inserted and any other online certificates that you might want automatically added to your resume like Mozilla Open Badges -> http://openbadges.org/

- The comment section suggests that lots of innovation could arise from it.

----------

Make sure you try the CLI! sudo npm install -g resume-cli

pserwylo• 12 years ago

I would just like to say that I think that this is a great idea(despite agreeing with others here that XML + XSLT would seem like a more logical choice). I hope it gets some traction in the long run.

I've posted an issue on github [0][1] mentioning that "firstName" and "lastName" is probably not a good representation of names. I don't actually know what the solution is, but this article [2] gives some great reasons why we tend to make incorrect assumptions about peoples names, which have the possibility of excluding some people form our tools based on their name alone.

[0] https://github.com/jsonresume/resume-schema/issues/4 [1] - It seemed like an appropriate place to comment. Please let me know if there is somewhere else you'd prefer such comments. [2] - http://www.kalzumeus.com/2010/06/17/falsehoods-programmers-b...

keehun2• 12 years ago

I agree with the names. Even just the ordering of first/last name can be difficult. I come from Korea where last name always comes before the first name in a formal setting.

Perhaps you should just require "full name" which the author of the resume would decide for him or herself.

yaddayadda• 12 years ago

Here's another good reference on international names and ways to deal with names - http://www.w3.org/International/questions/qa-personal-names

Uehreka• 12 years ago

Regarding names, I've heard of two solutions that cover enough edge cases to be suitable:

1. Asking for "Given Name" and "Surname" clears up most issues with cultures like Korea (where surname comes first).

2. Simply have a "Name" field. Accept any Unicode characters. Hope for the best.

If someone has a name that cannot be represented in written characters or cannot be rendered in Unicode (which would be amazing, since Unicode covers ancient dead languages and fictional languages like Klingon) they probably have more pressing issues than trying to serialize their resume.

dhawalhs• 12 years ago

I built something similar a couple of years back when I was job hunting using a json and Angular.js, I called it modern resume [1]. Live example can be found here [2]. I got tired of having a resume for each format - plain text, html (website), word document, PDF. Every change had to be made in all the places. I wanted to have a single source of truth. This kind of standardization would be great!

My resume.json for science [3]

[1] https://github.com/dhawalhshah/modern-resume [2] http://dhawal.me/ [3] http://dhawal.me/js/resume.json

aoxfordca• 12 years ago

Heyo another Tech alum :o) ... in good company

spb• 12 years ago

While we're showing our own resume formats, here's my own, half-YAML half-Markdown base: https://github.com/stuartpb/stuartpb.com/blob/master/routes/...

mitfik2• 12 years ago

I am very interested into your thoughts about LateX, why we need another "standard" if we have LateX? Do you provide something in addition which is not possible with LateX?

Thanks

chii3• 12 years ago

but is LaTeX machine readable - can i be given a latex document, and extract the essential data from it and transform it into some other format/structure that i can use elsewhere?

KingAndCaroline• 12 years ago

I've been working on a LaTeX-based resume builder (link in profile) and I could see something like this being useful for allowing users to take their data with them and move it between services. Anything that reduces the time it takes for people to input information and see the resulting image/pdf is worth looking into, so I'll be checking this out.

Karellen3• 12 years ago

You could use the "correct" spelling of the word: Résumé.

It took me ages to figure out what this actually was, because your non-standard "lazy" spelling made me think this was about resuming interrupted downloads/transfers of generic JSON data. (Possibly by doing a partial parse of the data that you have got, or over a non-HTTP protocol, otherwise why not just use HTTP "Range:" headers?) I had to unlearn that before the light-bulb went off.

bshimmin• 12 years ago

At least in the UK - and probably other countries in the Commonwealth - we almost exclusively say "CV" ("curriculum vitae", literally "course of [my] life"). "Résumé" is not well-understood here.

tallanvor• 12 years ago

Résumé, resume, and resumé are all accepted spellings in American English. As noted by others, many countries use the term C.V., which is likely to be a larger source of confusion for people.

johnchristopher• 12 years ago

Make sure you try the CLI! sudo npm install -g resume-cli

Regarding that: I recently reached the fed-up point with sudo or not sudo to install nodejs app and the mess it leaves behind when using sudo for just installing some local binaries.

I stumbled upon nvm[0] when installing docpad:

The above should not require sudo (the # means sudo). If it does, we recommend reinstalling Node.js so that it doesn't. Otherwise you're likely to run into permission problems in the future. Follow the Step 1 instructions to reinstall Node.js without requiring sudo.[1]

Which led me to this:

http://learn.bevry.me/node/install

and then this:

https://github.com/creationix/nvm

It makes npm installs everything under $HOME and never needs to be ran with escalated privileges. `nvm install X.X` and `nvm use X.X` allows switching between node versions and installed packages.

[0] https://github.com/creationix/nvm

[1] http://docpad.org/docs/install

vacri• 12 years ago

Your References fields need a section for contact details - email or phone. Testimonials are worthless by themselves.

Edit: references should also have an optional 'organisation' field and a 'position title' field. "This reference was the technical lead at Foo Corp, my second-last job. Contactable here"

pests• 12 years ago

I could see validators and other resume-specific checks coming about because of this.

Checking that dates are logical, spell checks, checking for dead links, showing employment gaps and so on.

thomasfromcdnjs25• 12 years ago

Great idea! Added as an issue to the resume-schema repo

krapp• 12 years ago

Random Project Idea - use this as part of an API which scrapes multiple sites (like Linkedin, Github, Monster, what have you) and builds a single, canonical resume from the various sources using this standard.

I don't know that the themes, necessarily, are that useful (I don't like the idea of implicitly connecting presentation with JSON like that - i'll make my own themes thank you) but I do like the idea of a common transport format for resumes.

thomasfromcdnjs25• 12 years ago

That sounds very ideal, I'd love to automatically insert my online achievements into my resume.

jimktrains22• 12 years ago

Now, if only people will actually use this.

jimktrains22• 12 years ago

To clarify, I meant that when looking for a job it is a pain to retype your resume multiple times (because having a human look at candidates apparently isn't a thing, especially at larger organizations). Having a standard format, supported by many employers would have made the process much easier.

My resume is set up as http://www.microformats.org/wiki/hresume and noöne supports that, and it's been around for a while.

bphogan11• 12 years ago

I think this is a great idea. I love the idea of a universal data format for resumes.

However, I'm not at all convinced that JSON is the way to do it. I think the way to do it is with XML and XSLT, although I already can hear the cries of "But XML is yucky and XSLT is hard." But by using JSON, we lose some semantic markup and end up having to parse it to create some semantic markup.

Still, I hope this, or something like it, takes off. This just needs a web-based GUI tool to publish a resume so that non-geeks can make it happen.

thomasfromcdnjs25• 12 years ago

It was a very hard decision to make and starting such a movement is unfortunately from my viewpoint much more than making rational choices about the technologies. We wanted to have a low barrier to entry and also take advantage of the large ecosystem of libraries. At the moment the majority of web technologies are using JSON so we believe that to get this standard off the ground we will need to go with the grain.

LinkedIn actually exports into a Microformat called hResume which has been around for a long time but has evidently not been able to amass a large developer community. This was another reason we decided to not start with XML.

xienze8• 12 years ago

LinkedIn actually exports into a Microformat called hResume which has been around for a long time but has evidently not been able to amass a large developer community. This was another reason we decided to not start with XML.

So, not trying to be rude here, but have you wondered why it is that an XML resume format hasn't taken off? Other than it being based on icky XML? I mean XML+XSLT has been around for a loooong time and can more than handle exactly what this format sets out to do, and yet a standard XML resume format hasn't taken off.

bphogan11• 12 years ago

We wanted to have a low barrier to entry and also take advantage of the large ecosystem of libraries.

That doesn't make sense. You wanted a low barrier to entry, but you have to mark this up yourself with JSON, install command line tools, etc. Seems to me a web app that I fill in some forms and generate a page on your site would be the quickest way to lower the barrier.

At the moment the majority of web technologies are using JSON so we believe that to get this standard off the ground we will need to go with the grain.

I understand the desire to use what's popular. But I just don't think it's the right choice. I'm interested to see the implementation of how this is turned into a PDF though.

Right now, though, it does sound a bit like "every problem is a nail." Perhaps it'll become more clear to me as I dig in more.

Again, I do think the project is very cool and will be encouraging my students to take a look at how they can use this to promote themselves.

esbranson3• 12 years ago

has evidently not been able to amass a large developer community

But let me guess: you will?

59nadir2• 12 years ago

At the moment the majority of web technologies are using JSON

The majority of which technologies? The CV generators that are already out there, which are the most relevant ones, don't use JSON. How are you creating a low barrier to entry by completely disregarding what people are already using?

crdoconnor8• 12 years ago

XML is a horrendous technology.

JSON was a much better choice. YAML would have been better still (it's more human readable), but both beat XML hands down in any situation.

bphogan11• 12 years ago

XML was horrible for data interchange, configuration, and all the other crap people used it for.

It's one of the best tools for marking up content. It's a Markup Language! If you've never used XSLT you really should look at what it can do. Because I can transform an XML resume into something gorgeous without writing too much programming logic.

It's the reason so many technical book publishers use XML. Easy to transform content while also describing that content.

bstar77• 12 years ago

Anyone that has ever used xslt in a non-trivial way most likely understands that it is a terrible answer to just about any problem. Complexity for the sake of complexity is never a good path to follow.

bphogan11• 12 years ago

I've written 7 books with it and those books have been published in print, PDF, epub, and mobi. I think that's non-trivial. But if you're referring to SOAP and WS* and all that stuff from the J2EE days, the I agree. That's a terrible terrible solution.

fs1112• 12 years ago

The EU has developed an XML standard for CVs years ago: http://interop.europass.cedefop.europa.eu/data-model/xml-res...

Qantourisc2• 12 years ago

year="1975" month="--03" day="---01" ???

icebraining2• 12 years ago

In my opinion, it should be RDF, so that we can programmatically define a common structure, link to shared concepts (e.g. programming languages) and to other personal documents. If it's encoded XML, JSON-LD, or (my personal favorite) Turtle, that's mostly irrelevant.

dheera2• 12 years ago

Nice. Although I'm not too sure what the social implications of this would be. As it is already I think employers do too much parsing of resumes rather than looking at them with human eyes. PDF resumes are awesome in that you get a blank sheet of paper to do whatever you want with to impress the employer.

There's a lot of reading between the lines that can be done in an resume. Even the fonts say something about the person. If the person is claiming to be a UI designer and their resume has poorly chosen fonts and bad kerning, I'd toss it. If they claim to be a physics researcher and the resume is written in Word, I'd toss it too. With a JSON resume you don't get any of this information.

Also, it's not entirely clear how extensible this "standard" is. Resumes take multiple forms. A computer scientist might want to list a few open source projects. A photographer might want to embed a sample or two of their work.

keehun2• 12 years ago

Yeah, I was going to say that this project is awesome, and I like the web themes, but the PDF export needs a lot of work. I think that will be the hardest aspect to achieve. Maybe some LaTeX resume themes can be borrowed from. There are some very nice ones in LaTeX.

dheera2• 12 years ago

Maybe. As someone who likes to be creative I'd still never use someone else's template though. Even the font I use on my resume I did some editing on because I didn't like the way a couple of the glyphs looked.

chillingeffect• 12 years ago

I'm glad you pointed this out.

The organization of a resume is almost as useful an introduction to a candidate as the terms on it. It's also useful to understand how much time was put into the resume.

If they claim to be a physics researcher and the resume is written in Word, I'd toss it too.

Hm, that's going too far, though!

dmaurath• 12 years ago

But the relationship with future performance and these subjective "reading between the lines" opinions is unclear. You need standardization to reduce bias in hiring decisions and make them as objective as possible.

est• 12 years ago

Need more formal ontological definition of the "skills" field, E.g. If you type "Django" the system should infer that you have fair amount of Python skills already.

Mister_Snuggles• 12 years ago

Some things that I think would be useful to add to the schema, in no particular order:

* It's common for resumes to include a full address, so elements supporting that could be added to the location. There should probably be a way to strip this out before publishing it to a public place though.

* I'd like to see a section for volunteer activities. I do some volunteering which, while not related to my day job, is worth highlighting.

* The references section would usually include contact information for a potential employer. Generally this would be a separate page that would be given during an interview. My main resume says "References available upon request" and I have a separate document with my references in it.

The biggest win, in my mind, is that the schema captures the knowledge of what belongs on a resume. Being able to take that single source and produce multiple versions (e.g., the public one, the private one with full details, a list of references, etc), not to mention everything else that comes with a standard schema, is a huge bonus.

thomasfromcdnjs25• 12 years ago

Great ideas! I will start add them as issues to the schema repo -> https://github.com/jsonresume/resume-schema

Your meta-layer idea is great too, at the moment we only have password protected resumes.

We are actually building an encryption converter, so you can encrypt your resume with a passphrase, I imagine these public,private etc versions could be controlled by encryption.

zak_mc_kracken• 12 years ago

Nice try, LinkedIn.

prat0318• 12 years ago

I have earlier written a similar json resume conversion gem - http://prat0318.github.io/json_resume I am currently working on i18n and a real time conversion web app. I would love to have something standardized and i hope it gets an active community.

sgonyea• 12 years ago

lickMyAsshole.json

kalleboo• 12 years ago

It took me until I got to the sample before I realized the page was talking about Resumés, not resuming file uploads or something. It was very confusing.

ladon86• 12 years ago

In the education section, I'd like to see an option for GPA or degree class. Be aware that some countries do not provide a GPA (for example, the UK has these classes: 1st, 2:1, 2:2, 3rd).

thomasfromcdnjs25• 12 years ago

Added as issue to the schema repo, thanks!

auvrw• 12 years ago

seemed like a good idea to me, although i didn't go the full 9 and write a schema or anything.

https://github.com/ransomw/ransomw.github.io/tree/master/med...

zobzu• 12 years ago

I find less-structured formats easier to read for this kind of data. Such as wiki-style markup (markdown, RST, etc.)

cm127• 12 years ago

Same here. My resume is a text file in markdown. Bullets, titles and subtitles are all you need for a simple, easy to read resume.

michaelq• 12 years ago

Would it make sense to mimic attributes present in LinkedIn's JSON structure? Most people already have a LinkedIn profile, and it would be easy to make a tool that exports LinkedIn profiles into you new open standard.

thomasfromcdnjs25• 12 years ago

You are correct, LinkedIn uses a Microformat called hResume, in our templates, we will export to HTML with built in support for the Microformat.

We also want to build a feature into the command line tool that lets users sync their resume.json to LinkedIn.

crdoconnor8• 12 years ago

If you want this to take off, you really need to provide a mechanism for exporting it to a wide variety of large resume handlers (monster, various large employers, etc.) and importing (e.g. from linkedin).

Otherwise you have the xkcd problem: http://xkcd.com/927/

It's boring and tedious work, and the number of edge cases will multiply geometrically, but it's likely the only kind of thing that would make this take off.

P.S. Why not YAML?

thomasfromcdnjs25• 12 years ago

We were nearly going to do a brand change because JSON and YAML are interchangeable and it doesn't really make any sense to just promote JSON in this case. I also love YAML, the problem was writing specification's seems a lot more mature when you follow the json-schema.org project. So I thought why not write it initially in JSON with the json-schema project as a basis and then just supply the tools for people who prefer YAML.

crdoconnor8• 12 years ago

Ugh, json schema is horrible. Have you used orderly? It's far less verbose, more readable and it compiles to json schema.

I agree that making it interchangeable is best overall (especially since there is a 1:1 convertibility between the two languages).

Perhaps the brand should actually reflect this instead. JSON is preferable in a data interchange scenario, but for keeping or displaying a human readable version, YAML is better.

As before, though, it's not YAML/JSON - it's the richness of the import/export filters that will make this idea float or die. Nobody much is going to want this so that they can publish their resume on jsonresume.org. People will want to use it so they can efficiently and easily import their linkedin resume and upload it to [insert fortune 500] company's custom HR portal / top 7 job sites.

BinaryIdiot• 12 years ago

This is a neat idea but this schema is a bit...specific in areas. Much like XML schemas of past the great thing about the JSON interchange format is that it's insanely flexible and anything goes (essentially).

I feel like it would be a much better way of handling resume type data if it was self describing rather than using a very specific schema that ties you into a data structure that may not make sense in many cases. In that case you could simply look for different types of data to display versus a very specific format that won't hold up to every use case.

jaltekruse• 12 years ago

While I don't jump for opportunities to fiddle with a word processor to make a document like a resume look good, I am curious if this is partially driven by an effort to populate databases with information. While I have been unimpressed by efforts to do this by scanning a doc or pdf file, I am afraid of the possible implications for encouraging poor recruiting techniques. Having a searchable database can be nice, but there are so many other layers to finding good talent.

steve918• 12 years ago

Seems like reinventing the wheel a bit here. Why not use jcard?

pavanred• 12 years ago

I keep thinking an ontology instead of JSON based resume would be a lot better.

oh_sigh• 12 years ago

I like the idea, because I deal with a lot of resumes and can't parse them programmatically.

Some thoughts:

- Eliminate first/last name distinction, and let people just put in their full or legal/passport name. Names are very crazy across cultures, as in some people don't have first or last names, so might as well just let them enter free form text as their name.

- Include a "Preferred name" field, also as freeform text. Some people like to go by nicknames, or shortenings of their names, and it will help out.

- Collapse location region and state into a single field.

- If "websites" is a hash with only a single field, call it "website" and make it a string.

thomasfromcdnjs25• 12 years ago

Awesome feedback, we already have an issue created for the firstName/lastName dilemma.

Will create issues for your other feedback -> https://github.com/jsonresume/resume-schema

awalton• 12 years ago

In other words, Microformats have arrived in the Web 3.0 sphere. (hResume never really caught on either, despite LinkedIn supporting it.)

baconchunks• 12 years ago

Another possible section to add: speaking engagements (or public appearances or perhaps the more generic "public events").

You have publications, but that implies written material is produced. And should publications have a place for co-authors? It seems wrong to not acknowledge them.

thomasfromcdnjs25• 12 years ago

Just added this as an issue yesterday -> https://github.com/jsonresume/resume-schema/issues/3

Thanks!

mindcrime5• 12 years ago

I have to admit, I'm having some trouble seeing the value in this... JSON does not strike me as a great format for a document of this nature, although it's certainly possible. Offhand, I'd think XML is a more natural fit for this.

Anyway, more to the point, there has been some existing work done on developing standards for representing resumes. I would highly, highly encourage you guys to look at this and consider making whatever you do compatible with the existing work as much as possible. Consider ResumeRDF[1] for example. Maybe consider doing JSON-LD and using the FOAF and ResumeRDF vocabularies?

[1] http://rdfs.org/resume-rdf/

DanBC• 12 years ago

I just had a look at ResumeRDF.

    Class: cv:Person

    Person - CV subclass of WordNet person

    in-range-of:	cv:aboutPerson cv:referenceBy
    in-domain-of:	cv:gender cv:birthPlace cv:hasCitizenship cv:hasNationality cv:maritalStatus cv:noOfChildren cv:hasDriversLicense

    [back to top]

It's probably a bad idea to encourage people to list protected characteristics on a CV / resume. Employers (except vary rare exceptions) can't use this information so there's no point including it.

mindcrime5• 12 years ago

That's an interesting point. And to be honest, I'm not sure what the rationale was for including those terms (I wasn't involved with defining ResumeRDF or anything).

I'm guessing it happened because they "borrowed" a generic notion of "person attributes" from some other, less specific, source, and didn't really consider the legal implications you mention.

I guess it's an open question whether or not there's any actual valid scenario where somebody would want to put that stuff. It seems unlikely to me, but maybe in other countries or something? Hmmm... dunno.

jerven• 12 years ago

In some countries this completely legal, and if it wants to be an international standard it should have space for it. (I hope there would be standard filters for each legal area e.g. filter out all protected characteristics for US applications)

akusete• 12 years ago

Looks cool, but Queensland is a state, not a region.

ttrbls• 12 years ago

lol

macspoofing2• 12 years ago

If only there was an extensible markup language you could employ to create a resume standard.

Seriously, JSON really sucks at markup (it also kinda sucks as data interchange as well, but let's not worry about that here). This 'standard' is so hand-wavy. Every property is a 'string', so why are you even bothering putting it in? Right, not really 'strings' because a bunch of fields have a JavaScript (so we're not really sticking JSON here, are we!) comments to describe it. Presumably order of object properties matters, or maybe it doesn't, or maybe it matters for some (e.g. root object / address) and not for others (eg. phone numbers). Date format? Who needs that! Which fields are optional? All of them? Some of them? None of them? I like the 'hobby' field. Complete waste of space on a real resume and a complete waste of space here as well (hah: hobby name and hobby keyword). Urgh.

It's a mess. A total mess.

thomasfromcdnjs25• 12 years ago

You are right, it is a mess, we are working on the specification at the moment and deciding on correct data types and semantics. If you think with a bit of love it can be made into a usable standard please come help us out at https://github.com/jsonresume/resume-schema/issues

macspoofing2• 12 years ago

JSON is the problem, so this will continue to be a mess. Especially if you have hopes for this to be used by many many people and especially if you hope this can be ingested by many many different systems (e.g. every resume submission server for every company). You need to formalize the structure, so you need some sort of a formal schema. You may pretend you don't need one, but you kinda have one right now. The schema you have right now is your example structure and JavaScript comments.

wldcordeiro• 12 years ago

Being an asshole on the internet isn't needed to get your point across.

grillp• 12 years ago

I know it's a little heavy weight, but there is already an open standard around this: http://www.hropenstandards.org/

They have a HR-JSON and HR-XML schema for HR docs..

Reinventing?

simonw• 12 years ago

It appears you have to register for an account on that site AND wait for your membership to be approved before you can even download a standard or view documentation.

If that standard is important and well designed, it should be trivial to write a converter from JSON Resume.

As Douglas Crockford once said (I believe in reference to XMLRPC) the great thing about reinventing the wheel is that you might get a round one.

indymike• 12 years ago

You can download without being a member. There is a Download Standards page that anyone can download from. It's pretty straightforward.

(on the HR Open Standards Board of Directors)

esbranson3• 12 years ago

Yep. I'm sure the fantastic exhibition of its fatal flaws will ensue. Maybe. Hopefully.

konstruktor• 12 years ago

As someone from a country with very different resumes, my suggestion would be to be more explicit about this being about US style resumes. If you want decent rendering, you need to focus on one style anyway (too much variation), so just be open about your scope and optimise the hell out of it, saving you a lot of i18n worries (names, addresses etc.).

stevesunderland• 12 years ago

this is f---ing brilliant.

sirji• 12 years ago

Gravatar Url?

thomasfromcdnjs25• 12 years ago

We should be able to generate those from the email. Good suggestion though, I'm going to add mine to my resume.

leed25d• 12 years ago

Thanks, but I'll keep using LaTeX.

treve• 12 years ago

I think you missed the point. It's not about displaying and formatting resumes, it's about exchanging data between software.

korzun2• 12 years ago

As somebody who deals with hiring, who is going to use this exactly?

The only 'benefit' I see is that the crawlers/recruiters can parse your resume easier.

Great for them; usually not so great for you.

lylejohnson• 12 years ago

Are you kidding? With this, I can more easily skip over those applicants with tons of Java and C++ development experience when I'm looking for a C# developer. No more wasting time on reading resumes with completely unrelated experience!

korzun2• 12 years ago

Why would I be kidding?

You are just backing up my point that recruiters will use it to grab your data and filter you like a product.

There is absolutely no benefit to you as an engineer. Quality / noise ratio is the issue here.

esbranson3• 12 years ago

Europass CV (can be embedded in PDF files, easily exports to everything using EU government web services):

http://interop.europass.cedefop.europa.eu/

HR-XML (also comes in a JSON variant, "HR-JSON"?):

http://www.hr-xml.org/ http://www.hropenstandards.org/

There is also Microformat (hResume):

http://microformats.org/wiki/hresume

Europass is used by the European Union, HR-XML is used by the US government. Microformats are probably used by many. This stuff is OLD news. We certainly don't have a shortage of usable standards.

chii3• 12 years ago

it's a classic case of http://xkcd.com/927/

fs1112• 12 years ago

but this one is written in NodeJS. So it is so much more scalabale!!!11!!!! :-)

iaskwhy• 12 years ago

Other people might be able to confirm it but I have never received an Europass while working in London. I was told I should never send a cv made with the Europass template as the relation between British people and European things is not always the best. I understand this might be incorrect but it was enough for me to avoid using it. Then I also spoke with some HR people from other European countries who told me they didn't really like Europass so in the end I created my own template using Bootstrap.

Argorak• 12 years ago

Europass is also largely unknown. Politics aside, it is a good format and I would much prefer to get standardized CVs like this, especially for the skill assessment in languages.

The problem here is: some European countries use very particular formats for their CVs. e.g. Germany has the "tabular CV", which expects all these things to be organized by time and date, it even includes where and when you went to Kindergarten. Europass doesn't fit many of those particularities (which is kind of the point).

toyg• 12 years ago

I had a brief look at HR-XML back when I was in love with XML - it's a very old standard, early-2000s, and yet it's never mentioned by any recruiting website, even "geek-oriented" ones that should know better.

At one point I even added an explicit line to my Word/PDF versions: "this document is also available in HR-XML format on request". Nobody ever asked, so eventually I just took the line out. I've never heard of Europass CV, but I live in England, where EU is a dirty word...

TBH, this is the sort of stuff most recruiters don't bother with anyway, in my experience the (European) recruiting industry runs on MS Office. Some industry vendors might do some munging and conversion behind the scenes, but they'll likely never allow people to submit their own data outside of tightly-checked web forms.

ARussell• 12 years ago

I disagree with so many people in this discussion recommending XML without much reason. The best reason I've ever seen to pick XML over JSON is mixing text and elements, and that doesn't seem like an issue, here.

As for feedback on the schema itself, I'd probably like to see some more detail allowed for the courses under education. I remember after graduating, there really wasn't anything more substantial that I could put other than details about my courses. Using the example on your page, I think "HS1302" and "Introduction to American History" could possibly be separate fields, and I think there should be an additional details/description/highlights field.

Also, I am a consultant, and I like to list a projects section. Each project could have a name, a client, an employer (the consulting company), a date range, and details.

thomasfromcdnjs25• 12 years ago

Thanks! Will copy your feedback where appropriate to the issue tracker

b3n• 12 years ago

Could this be used with Namecoin (where JSON values are stored in a blockchain)? Perhaps something like https://github.com/onenameio/onename

tommorris2• 12 years ago

That'd be a great way to make a pointless format even more pointless. But, sure.

alexatkeplar3• 12 years ago

This is cool! We do a lot of work with JSON Schema at Snowplow. Here's a few suggestions:

1. Semantic versioning doesn't work for schemas. (What's a patch/bug fix for a schema? It's meaningless.) Use SchemaVer instead - http://snowplowanalytics.com/blog/2014/05/13/introducing-sch...

2. Make your schema self-describing so that a given resume instance is associated with its schema version http://snowplowanalytics.com/blog/2014/05/15/introducing-sel...

3. Publish your JSON Schemas to Iglu Central, the first schema registry for JSON Schema - https://github.com/snowplow/iglu-central

thomasfromcdnjs25• 12 years ago

Adding your issues to the tracker and also thanks for the link to Iglu central!

alexatkeplar3• 12 years ago

Nice one! If I can help in any way, just let me know. Iglu and Iglu Central are very young but we're v excited about them.

nileshtrivedi• 12 years ago

Nice. Now create a schema for job postings as well so that one can write a crawler to find all interesting vacancies in one's favorite companies.

bitL• 12 years ago

Ugh, what's the point? Hiring by robots?

kijin• 12 years ago

Looks cool, although it looks a bit U.S.-centric and techie-centric.

Other people have mentioned the firstName/lastName problem and the region/state ambiguity. I'll mention a few more items that might be helpful if you really want this to become a worldwide standard:

1. In some countries, it is not only legal but also customary for employers to ask for, and job seekers to include, various demographic information about themselves that would be considered discriminatory in the U.S. For example, sex/gender, date of birth, marriage status, and family details. It would be good to define such fields in the schema but leave them optional. Or define a catch-all field where anyone can add any key-value pair.

1-1. Don't forget the full street address, including the postal code!

1-2. Citizenship. This is very important for companies that might need to sponsor your visa and/or geeen card.

1-3. In some countries, it is customary to include a photograph of yourself in your resume. Perhaps this could take the form of a "photo" field that takes a URL.

1-4. There needs to be a field where you can write down an alternate expression of your name. Many Korean names can be written in either Hangul or Hanja. Most Japanese names are written in Kanji, but often accompanied by Hiragana to help the reader pronounce them properly. A lot of employers in these countries require both forms.

2. Since this is going to be electronically processed, perhaps it would be a good idea to include the ticker code for publicly traded companies in one's employment history? GOOG is much easier to search for than variations of "Google", "Google Inc.", etc.

3. The publications schema will need to be significantly expanded if you want scientists and other academics to even consider using this. Take a look at how academic publications are cited. At the very least, there needs to be a field to specify the type of publication, e.g. book, ebook, edited book, journal article, newspaper article, conference presentation, workshop presentation, blog post, website, etc. There should also be a "doi" field for scientific and other academic publications.

3-1. The education schema needs a "summary" field, just like the employment schema.

4. Some employers prefer to get the content of the reference from the referee him/herself, rather than having the applicant include it in the resume. (Otherwise, the applicant could modify the content.) So you need fields to put the email address (to contact the referee) and a URL (to grab the content of the reference) in the references field. Oh, and don't forget the referee's employer and title/position. Those things matter.

4-1. Bonus points if the content of the reference can be authenticated with the referee's PGP key.

4-2. Extra bonus points if the entire resume can be authenticated with the applicant's PGP key, to prevent hiring agencies from messing with the content.

thomasfromcdnjs25• 12 years ago

Excellent feedback and you've set the bar high which is good! I don't have the time to reply to each point at the moment but will definitely copy it over to the issue tracker.

Qantourisc2• 12 years ago

My main concern and same question for my own resume: what skill strings should be used ? Server Administrator or System Engineer ? Programmer or C++ ? etc... That's what will be one of the more challenging things to solve. Otherwise you might have a resume-format, but no way to reliably use that data as a computer, or that out of scope ?

icebraining2• 12 years ago

This is why the Semantic Web has links (URIs), so that we point to unequivocal concepts. They would do much better to use JSON-LD instead of dumb strings for these CVs.

thomasfromcdnjs25• 12 years ago

We are struggling with this part, as the other reply stated our strings are bad but they are also useful for lowering the barrier to entry, we are hoping to find a sweet spot in the middle somehow.

TeeWEE• 12 years ago

This is exactly where XML is the only right choice. JSON is much simpler than XML, thats why lots of people like it. However for documents XML markup is more appropriate. The reasons:

- Allows extensibility with other xml standards (such as SVG and MathML in your resume, which might makes sense) - Allows linking to other stuff with XLink - Allows styling it with lots of xml styling technologies. - Better versioning support

JSON is not bad however, so if you just want to have simple plaint text cv's, json is good enough. I do agree that with XML comes bloatness, complexity, and more crap. However from a pure technical point of view, its a better option for doing resumes.

stevekemp• 12 years ago

I have a resume I've been working with for a few years now, and it was built using:

http://xmlresume.sourceforge.net/

That project seems dead now, and recent updates to my systems have resulted in me being able to build HTMLTXT but not PDF.

In the future, when looking for a job, I'll be looking at JSON instead. It might not be the best tool, but it seems to be what several _live_ and active projects are using.

peteretep• 12 years ago

I write software that interacts with CVs/Resumes for a living. Why would I use this over something based on HR-XML?

emiller829• 12 years ago

This reminds me a bit of an idea I messed around with a bit last year: http://erniemiller.org/me.json

I posted about it then (http://erniemiller.org/2013/06/19/my-happiness-formula-me-js..., but the thought at the time had been that it would be nice if prospective employees published something along the lines of things that contributed to their contentment with work.

This was eventually intended to be presented sans ratings in a little browser plugin, for recruiters and the like to score themselves honestly against the criteria, and a little bit of math would yield a compatibility rating on a click.

It wouldn't take long until such a thing evolved into a more full-featured resume such as this. I like the idea.

Xophmeister• 12 years ago

Beginning to wish I hadn't abandoned my modular JSON to LaTeX CV builder... C'est la vie!

thomasfromcdnjs25• 12 years ago

ahh that would have been perfect!

ironlady• 12 years ago

Very cool! I started something like this with a WordPress implementation a few years ago and never bothered to complete it, so cool to see somebody else with a similar idea and actually finished it!

codgercoder• 12 years ago

Doesn't it ever dawn on anyone that employers insist on detailed, accurate information about applicants, in standardized forms, while the information available about employers is a mess? There's no standard, there's no requirement, and they can mislead in quite creative ways. Hardly seems fair.

ubershmekel• 12 years ago

http://www.glassdoor.com/ is a good start.

n0body• 12 years ago

stupid

exabrial• 12 years ago

an XSD?

edoceo• 12 years ago

Another multi format resume/CV tool: http://ars.io/

yellowapple• 12 years ago

Looks nice (and this is coming from someone who's pretty weary of the "hey let's put the whole world on npm" attitude). Seems to be a pretty decently-thought-out structure; after converting to YAML (my preference), it's pretty straightforward to work with.

Some comments:

- I found it a bit odd that 'resume init' only prompted for firstName and personal email; a more complete wizard would be useful. I suppose it wouldn't be hard to create a third-party "JSON resume generator" app with a nice friendly pretty interface for this (it is just JSON, after all).

- As mentioned in other comments, the "firstName" and "lastName" fields make excessive assumptions about the user's native/preferred naming convention.

- The schema only provides for a "blog" element in the "websites" array. More options would be appreciated, since not every personal website is a blog. Alternately (if the intent is really for only one website entry), providing a scalar/string "website" element would be more to-the-point.

- As someone who prefers YAML (it looks so much nicer), built-in support for YAML<->JSON conversion would be awesome for me (i.e. be able to export to / import from YAML). For now I'm fine with using yaml2json/json2yaml, but such import/export functionality would be awesome. Perhaps a command-line switch for exporting (i.e. "resume init --yaml")? Alternately...

- On the above note, would it be possible to make "resume init" output to STDOUT instead of directly to a file? That way, it would be simple to run "resume init | json2yaml > resume.yaml" (for example).

thomasfromcdnjs25• 12 years ago

Thank you! We are inundated with work right now so I can't reply fully but I will probably end up using the YAML version myself so I've created an issue to implement YAML into the CLI -> https://github.com/jsonresume/resume-cli/issues/20

Will add your other feedback to the issue tracker soon.

hunvreus• 12 years ago

I've been thinking about exporting my resume to a similar format (hResume or equivalent) for a while; I find LinkedIn of little value for me.

I can imagine that for certain people it offers opportunities to "network", but in my case it is more an invitation for random people to spam me than a tool to connect.

Beyond the resume, I am wondering if it would be possible to build some kind of decentralized professional network, with resumes as nodes and professional/academic relationships in between, maybe even a vetting system (LinkedIn recommendations are not only time consuming but also worthless IMHO).

59nadir2• 12 years ago

Is it just me, or does it feel just a tiny bit ridiculous to see Twitter proposed as being on a CV? GitHub, sure, but Twitter? Shouldn't we just add our WoW accounts and maybe our old ICQ numbers too?

For people interested in standardized CV, by the way: http://europass.cedefop.europa.eu/en/documents/curriculum-vi...

This site details some stuff about working with Europass: http://interop.europass.cedefop.europa.eu/

I'm surprised they didn't go for interoperability with what's already out there.

jaseemabid• 12 years ago

I'm happy that there is no 'Career Objective' section. This website[1] is down now but I think the name convey the idea.

1. http://whatthefuckismycareerobjective.com/

mb2100• 12 years ago

I just wrote my resume in markdown and export it to txt/html/doc/PDF with Pandoc: http://johnmacfarlane.net/pandoc

tommorris2• 12 years ago

Just use HTML + hResume. hResume is based on vCard and thus the name and address fields are actually sensible rather than hacked together without considering edge cases. It lets you use prefixes and suffixes (like Dr, Prof, Rev, PhD, MD, JD etc.). It lets you format text using HTML tags. You can link to it.

If you want JSON, you can export it from the hResume using a parser.

Unlike JSON (or XML), non-programmers can read HTML in a crazy new piece of technology called a web browser without having to run it through any transformation tool. Want PDF to send to a company? There are a few WebKit PDF renderers available.

Want to style it? CSS.

Want nice fonts? WebFonts.

It's a web page. Crazy idea.

Classic version: http://www.microformats.org/wiki/hResume

New version: http://www.microformats.org/wiki/h-resume

pacmanche• 12 years ago

Created a JSON Resume LinkedIn exporter https://github.com/mblarsen/resume-linkedin

sktrdie• 12 years ago

There's already a W3C Ontology for Resumes: http://www.w3.org/wiki/ResumeRDFOntology

All you have to do is use the terms with JSON-LD.

ff7c11• 12 years ago

http://microformats.org/wiki/h-resume looks good as you can choose how you want to display the data as well as allowing parsing to JSON if you want

troydm• 12 years ago

I've been more partial to schema.org's JSON-LD patterns. I'm curious what key differences made you opt for one over the other.

I could see this as a natural expansion of http://schema.org/Person

JSON Resume – a JSON-based open source standard for resumes

Related Stories