Spark – Frequently asked questions

This page includes frequently asked questions for the Spark browser extension. You can download Spark via the Chrome Web Store.

Can I use this extension as a reference manager?

No, you shouldn’t use Spark as a reference manager. It does not create backups of your history and you cannot manually add research outputs or other sources to it. It operates automatically by saving DOIs that it detects, so it is more limited in functionality compared to a reference manager.

How does the extension find a DOI while I browse?

The extension works in the background as you browse the internet. DOIs are typically available in dedicated HTML tags or as text within a webpage. Spark is able to spot DOIs, which are then used to provide you with a formatted citation, a free-to-read version of the article and cited materials, where available.

Are there versions for other browsers, like Firefox or Safari?

At the moment, Spark is only available on Google Chrome and Chromium browsers. This is simply because the extension is in its very early days, so it was developed for the most popular browser. If there is a lot of interest in it, it might become available on other browsers, too.

What data does the extension store?

Spark stores a DOI history based on the articles you view, plus your saved settings. These include an email address and a CORE API key (if you decide to enter one). This data resides in your browser and gets synchronised between Chrome installations (e.g. Chrome on your laptop and Chrome on your desktop). However, the data always stays with you and is not sent to any third parties/advertisers/etc.

Note that Spark makes API calls to Unpaywall, CORE and the DOI Citation Formatter, so DOIs, an email address and a CORE API key will be sent to these services as appropriate. Just to clarify, nothing shady is happening – sharing this information is perfectly normal and you can see how API calls work in the documentation for Unpaywall, CORE and the DOI Citation Formatter.

Why should I enter my email address or an API key?

Unpaywall requires API calls to include an email address. This is set by default in Spark, but you’re better off entering your own: if Spark gets a lot of interest and everyone uses the default email address, it will hit Unpaywall’s daily request limit. The CORE API key is simply essential to make API calls to CORE. If you don’t enter one, Spark will not be able to return open access content from the CORE database.

Can I use the code behind the extension to build my own project?

Definitely – knock yourself out. The code uses a GNU GPLv3 licence and is available to download via GitHub.

Please keep in mind that I’m not a professional developer. If you have any suggestions to improve the code, please contact me at chiarelli [dot] andrea [at] gmail [dot] com.

The extension doesn’t work on a website where I can clearly see a DOI. Why?

That’s unfortunate! I have tested Spark on lots of different websites, but some might have slipped between the cracks. Please contact me at chiarelli [dot] andrea [at] gmail [dot] com with the URL of the page(s) where Spark didn’t find a DOI and I’ll see how this can be fixed.

I can see the “Cite” button but nothing happens upon clicking. Why?

You should check whether the website is using Https: unsecured websites don’t let Spark copy into your clipboard, so you might see the “Cite” button but this won’t work. If the website is not using Https, I recommend you install the extension “HTTPS Everywhere“: the extension encrypts your communications with many major websites, making your browsing more secure, and also fixes this incompatibility with Spark.

Some of Spark’s buttons are greyed out and won’t do anything. Why is this happening?

Spark will display all three buttons when it detects a DOI (“Cite”, “Read” and “Explore”). However, if no free-to-read version or cited articles are available, the respective buttons will be greyed out. This is meant to show you that Spark has made its API calls and looked for the information as intended, but hasn’t found anything.

If the “Cite” button does not appear, it is very likely that Spark has found a DOI but this includes some textual elements (i.e. gibberish) that aren’t recognised. If this happens, please contact me at chiarelli [dot] andrea [at] gmail [dot] com and send me a link of the page where the “Cite” button was missing.

Does the extension work with books?

It depends. If a book has a DOI, it does. However, if a book only has an ISBN, it doesn’t. This feature is quite complicated to add when books don’t have a DOI, as databases with structured metadata on books are not as mature as the ones for academic articles.

I can’t see my favourite referencing style. Can it be added?

Definitely. Please contact me at chiarelli [dot] andrea [at] gmail [dot] com and let me know what you’d like to see on the list. Generally speaking, I would not add institution-specific styles, but I am pretty open to everything else. You can find a full list of options in the dropdown menu titled “Select Formatting Style” on this page.

I know of an amazing API that works with DOIs. Can this be integrated in the extension?

In principle, I’m open to doing more with Spark. Please contact me at chiarelli [dot] andrea [at] gmail [dot] com to share your idea(s) and we’ll take it from there!

I would like to share and/or promote this extension. What should I do?

Please contact me at chiarelli [dot] andrea [at] gmail [dot] com and we can talk about this. I am open to sharing Spark as widely and broadly as possible.

I have lost my browsing history in the extension. Is there anything I can do?

Unfortunately not. The history is only saved on your device(s), so there are no backups elsewhere. This is why you should use a reference manager for any resources that you wish to keep safe in the long-term.

I’d like to export my browsing history and put it in a reference manager. How do I do that?

You can do this in two main ways:

  • Option 1: A simple copy and paste (works with Zotero). Open Spark’s ‘History’ using any referencing style that shows a DOI (e.g. APA). When the history is shown, just click the “Copy all references” button and paste them into the ‘magic wand’ tool in Zotero. When you press ‘Search’, all DOIs should be detected correctly and added. This approach may well work with other reference managers – I just haven’t tried!
  • Option 2: Using BibTeX. Open Spark’s ‘History’ using the BibTeX referencing style. When the history is shown, just click the “Copy all references” button. BibTeX references should be broadly compatible with any reference manager, and you can paste them into a text file created using the Windows Notepad: if you save this file as ‘whateverName.bib’, most reference managers will let you import this.

How does Spark make money?

The answer is simple: it doesn’t. It is currently a personal and completely not-for-profit project, so you can rest assured that your data won’t be sold to the highest bidder and is only available to you (or whoever has access to your devices).