Bug 791065

Summary: Spelling error in virt-manager GUI
Product: Red Hat Enterprise Linux 6 Reporter: Laura Bailey <lbailey>
Component: virt-managerAssignee: Cole Robinson <crobinso>
Status: CLOSED NOTABUG QA Contact: Virtualization Bugs <virt-bugs>
Severity: low Docs Contact:
Priority: unspecified    
Version: 6.1CC: bforte, dallan, hjiang, mzhan, rwu, yupzhang, zpeng
Target Milestone: rcKeywords: Reopened
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-03-09 21:20:49 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 753997    

Comment 2 Huming Jiang 2012-02-24 08:33:42 UTC
I tried to find the word 'paravirtualize' on virt-manager GUI of rhel6, but without result. Maybe i missed something. Could you please tell me how to find it?

Thanks very much.

Comment 3 Cole Robinson 2012-03-02 21:09:23 UTC
Thanks for the report, however "paravirtualization" seems to be the more popular of the two, yeilding 4 times the number of results on google, including on intel and vmware sites. The wikipedia article for paravirt doesn't have a dash either. Granted spelling isn't by consensus but I think the common usage is without the dash. Original libvirt documentation doesn't use the dash either, nor the qemu man page.

I think changing this in virt-manager would be inconsistent with most of the rest of the virt world. Closing as WONTFIX, please reopen if you disagree.

Comment 4 Brian Forte 2012-03-05 04:25:30 UTC
The word ‘para-virtualise’ (and variations thereof, such as para-virtualisation) hasn’t been captured in any of the corpora I can find, including the Oxford Corpora, which is the primary research source for the OED.

Treading the murky waters of search-engines*, all four spellings you’d expect (given Commonwealth and US English differences) abound:

    1. paravirtualise

    2. para-virtualise

    3. paravirtualize

    4. para-virtualize

Unfortunately, search-engines are bad indicators of regional use, since their algorithms play merry hell with geography.

Also, at least so far as Google, Bing and DuckDuckGo are concerned, our own documentation is a key source for this word (and we’re not spelling it consistently).

IBM and Microsoft have plumped for ‘para-virtualize’ (the US “‘-ize’ is the only ending” pattern plus a hyphen).

Our own docs appear to generally follow this as well.

In search-engine land, spelling 4 above (the one used by IBM and Microsoft), is not the majority spelling.

In formally edited and published documents other than our docs, IBM’s docs or Microsoft’s docs, the split appears to be about even between hyphen and no hyphen. VMWare’s docs and Citrix’s docs, for example, mix the hyphenated and non-hyphenated version roughly evenly.

Unsurprisingly, all these formal sources use the ‘-ize’ ending.

In the absence of better data†, I’d recommend spelling 4 above: para-virtualize.

IBM uses it. Microsoft uses it. And we appear to have plumped for it for the most part as well.

Given time, and the en-US tendency to swallow hyphens, I’m confident ‘paravirtualize’ will eventually become the standard spelling.

That time hasn’t arrived yet, however. And, since we write formal English‡, we should approach spelling and usage trends conservatively.


* Usage note: when using search engine results as a data source, I
  always search from within an anonymised Tor connection and use HTTPS
  as the connection protocol.

  Not out of an overweaning paranoia, but to obviate the strong
  tendency of all such engines (Google’s and Microsoft’s in
  particular) to personalise and contextualise the results based on
  previous searches done using the same browser. This can skew results
  badly.

  When I query Google from my day-to-day browser, for example, my
  searches on ‘para-virtualisation’ (and its variant spellings) are
  strongly skewed towards our own documentation.

  This is not surprising given my search habits (I routinely search
  docs.redhat.com using Google and the site: prefix) and browsing
  history, both of which are used by Google and Microsoft to
  personalise my results.

  Even given this, search engines remain a poor substitute for
  corpora in establishing real-world trends and usage.

† I’ve opened a ticket in the Oxford Corpora noting the word’s not
  being captured effectively, so I hope we’ll have data relatively
  soon.

‡ Not stylised English, formal English, aka standard written English.
  It’s less about tone and more about word, editorial and presentation
  choices.

Comment 5 Dave Allan 2012-03-09 21:20:49 UTC
Thanks for the exhaustive treatment of the subject.  Red Hat has a problem in this area stemming from our FOSS roots, which is that it's not economically feasible to patch all the products we package to abide by a particular region's tradition, so IMO we have to declare this kind of situation not a bug by fiat and trust that our customers will understand.