On Wed, May 29, 2013 at 9:08 PM, Graham Lauder <[email protected]> wrote: > On Tuesday 28 May 2013 21:07:28 Rob Weir wrote: >> On Tue, May 28, 2013 at 8:27 PM, Graham Lauder <[email protected]> wrote: >> > On Tuesday 28 May 2013 15:00:47 Jürgen Schmidt wrote: >> >> On 5/28/13 2:48 PM, Alphonso Whitfield III wrote: >> >> > I agree with Kadal looks like more of the same. >> >> >> >> And that is not necessarily bad. We got feedback from 5000 users and it >> >> seems that the majority like the logo we have and why not simply keeping >> >> the main idea and do only some refresh. >> > >> > Where is that data, I can't find it, it doesn't seem to be on the wiki >> > with the logo stuff >> >> Hi Graham, >> >> Thanks for checking in. I'm sorry you were not able to be involved >> earlier over the several months that we've been working on a new logo. > > Likewise, oh well those are the breaks, priorities must needs win out. > > >> We had 40 proposals, did a survey and now we're refining the highest >> scoring submissions. If you want to get caught up on what we've all >> been working on, a good start would be the blog post here: >> >> https://blogs.apache.org/OOo/entry/results_of_apache_openoffice_4 >> >> Note also the link to the more detailed report: >> >> http://survey.openoffice.org/reports/aoo40-logo-poll/ > > Excellent, thank you. Why were Lucas Filho's concepts not included? > > https://cwiki.apache.org/confluence/download/attachments/27846912/Concepts_LucasFilho02.png >
Human error? Though the survey was vetted on the dev list and ample opportunity given to object if anyone thought something was missing. In fact some other logos were missing and the designers noticed this and told me. The process certainly is biased toward the interests of active community members. > > > >> >> > When you say "Not necessarily bad", by what criteria do you judge this, >> > the >> > aesthetic sensibilities of 5000 random respondents. I'm sorry I missed >> > the >> > survey so I have no idea of what was in it. >> >> We prefer to call them "users", just like we call those who comment on >> the list "community members" rather than "random posters". > > Without a specific demographic defined and no filtering on respondents then > they > are to all intents and purposes random. You could define them as "Community" > but concievably anyone who owns a computer could be called part of that > community as soon as they make an effort to interract with the project. > > The only non-random element is that they all knew about OO before doing the > survey which pretty much puts most of them outside the demographic that we > need to reach. > Fortunately we also included demographic questions so we can filter results and look at differences in a more fine-grained way, e.g., long term OpenOffice users versus those who are not users, by gender, by age, by country, etc. > >> >> > People like it because it's familiar, however it fulfills none of the >> > requirements of a brand relaunch. I doesn't signal a new beginning, just >> > says "same old, same old". Besides which, choosing a logo by "Vote" is >> > going to purely subjective without any thought being given to the >> > practical needs of the brand, especially that of the marketing side of >> > things, without any thought to the Brand as a whole. >> >> Since the respondents to the survey were primarily current OpenOffice >> users it is natural that the results would be biased toward >> continuity. > > Indeed, so therefore from a marketing POV the survey/poll was meaningless in > terms of the branding needs. A number of us (Not just me as you seem to be > inferring later in this reply) stated early on, that a Vote is a bad way to > select a brand. As I probably pointed out back then, we produce a consumer > product. Our users are not developers or techs and the "User Community" is far > larger than the "Project Community". If we were to make a comparison with the > High Street then we are a B2C operation rather than a B2B which most other > Open Source projects, certainly those here in the ASF, tend to be. > We did not vote on a logo. We had a preference polll to gauge perception on the proposals from ordinary users. This data was gathered objectively, analyzed objectively and the analysis is being used objectively. How the logo appears to "random" people is relevant information. The comments especially are relevant. It is good that we know which logo is too close to the emblem of a European political party, which one is derived from an Icelandic radio station, which ones are unlucky in Chinese, and which one looks to some like a "flying penis". Without the survey these relevant items might have been missed. But in no case is the survey treated like a vote and determine the outcome by itself. It is just an additional data point to feed into a deliberative process. If it was a vote we'd be done by now. > >> >> On the other hand "continuity" is a value just as "new beginning" is a >> value. > > Definitely, I agree that continuity is a value but mainly in a shrinking or > static market. We however are in an expanding market and to retain relevance > in a growth market requires a brand that is fresh and dynamic. > >>There is nothing wrong per se with expressing continuity by >> having a logo that is a more modern take on the classic logo. >> >> . > > If that were true, then the orb would be gone. The Orb is not "classic", > Stella knocked that one up as part of a refresh post Oracle takeover. It's > not classic in fact it just reminds me of the whole Oracle screw up. > The "Classic" logo as I have pointed out in the past ad nauseam is 14 > characters in Camel case: OpenOffice.org, THAT is not available to us. The > truly > recognisable classic part of our branding has been stripped from us by short > sighted Apache policies. > The gulls and the color and "OpenOffice" (camel case even) are parts of the continuity. > >> >> <snip> > >> >> > The PMC needs, like any good management team, to ignore the subjective >> > and >> > >> > make a decision on purely objective criteria: Impact, uniqueness, story. >> > >> > If this was a decision being made by a marketing department run by me, the >> > toss up would be between Lucas's "Warpaint", his feather in orb, Kevin's >> > feather in Orb but with perhaps a more quill type motif. I also would >> > look at Robin Fowler's feather motif in 16 as a quill, it's simple shape >> > has scaleability advantages. >> >> It might make sense for you to express a preference for a single logo, >> or submit an alternative choice, and then argue for your "objective" >> criteria rather than merely asserting that you along are gifted with >> true insight here and everyone else is merely blindly following >> personal subjective taste. Remember, each of us could call our views >> objective as well. But where would that get us? > > > Heh, not everyone else, there have been other voices of reason, but they're > probably too polite and are easily ignored. Those people with marketing > knowledge on this list are few and far between and thus are easily drowned out > by the clamour of those who see marketing as "Puffery" or just an annoyance. > Noted: everyone that agrees with you is silent. > The Linkert scale is purely about subjectivity so any response generated by > the survey is going to be subjective, that's not an observation, that is fact. > It is an objective measure of the subjective perceptions of respondents. Public perception of the logo is a relevant fact. > From a marketing perspective, and purely about brand recognition if you were > going to use the linkert scale as a guide then take out all the middle > responses and throw them away. For our purposes they are meaningless. They > are a non reaction, in other words boring. You would be left with those that > excited a reaction whether negative or positive. > This was in the report, in this chart, though looking at only Strong Likes: http://survey.openoffice.org/reports/aoo40-logo-poll/strong-likes.png (I don't think the Strong Dislikes take us in the right direction.) > The data is then revealing: For instance the respondents are negative types > making 2.5 times more strong dislikes votes as strong likes. The logo that > generated the most strong likes (No4) only got around ten percent of > respondents voting for a strong like. That tells me that few of these logos > have any of the sort of impact that would be needed to generate significant > increase in brand recognition. > > At the other end of the scale however No37 generated a strong reaction in > close to 50% of respondents. That's what I call significant. However, it's > red and orange, colours which are known to generate reaction and high > interest. 37 was also square so it was more in your face given the > previously mentioned inequities with regard to image sizing. Given also the > demographic of the respondents it was also the one that was most unlike the > present logo so there was always going to be a bias, as you've already noted > above, toward the encumbent and also likely to generate the most negative > response. > 37 had the largest number of Strong Dislike scores. The comments on this logo were very negative as well. It would certainly be noticed, but I'm not sure that the associations would support the brand. > Most significantly however our present logo seemed to generate a "shrug" level > of enthusiasm being 35th on the list for generating a response. Very much a > Ho-Hum response, with only around ten percent of respondents registering a > strong reaction whether Like or Dislike. This means our Brand doesn't even > excite our own community which is the most significant indicator for the need > for change and in fact in just under 4400 responses our present logo only > generated 250 odd more Strong likes than 37 which has the most strong > dislikes. > I think we want to consider the base rate of responses and look at deviations from that. In any survey with a neutral middle choice there will be a tendency for respondents to pick that. I did another charts, not included in the report, to eliminate the neutral scores, looking instead at (Like + Strong Like) - (Dislike + Strong Dislike), but the results don't change much. > > Marketing's main goal should be to increase brand recognition amongst users > outside the present user base. To do this, one thing we should do is excite > our present users, word of mouth is the best form of marketing. If this > survey highlights anything it is a disturbing level of indifference. > Note that the report does show the scores of those who are OpenOffice users versus those who are not: http://survey.openoffice.org/reports/aoo40-logo-poll/users-notusers.png As noted in the report there were some logos that placed higher in one group or another, but at the top of the scale the preferences were stable. I do agree that the primary audience is for those who are not yet users. And if there were large differences in that chart then I'd be more concerned. But the data doesn't show a difference here. > >> >> In any case, I'll be sure to put an "other" choice in the ballot for >> the vote, so you can express choices other than the ones on the wiki >> currently. > > No real point, my preference is based on different criteria to the others, the > survey used a linkert scale which is designed to be purely subjective. > As has been noted before, a vote is the worst way to choose a brand because it > will always tend towards the bland. No point in contributing to a flawed > process. > Fortunately we did not vote on a logo, as explained above. The PMC will pick the logo, and the results of the user preference poll is one data point for their consideration. Your views, your opinions, of course, are also input into the process. Regards, -Rob > Also I have in the past noted on this list the criteria that good branding > needs to fulfill, at this point none of the above fulfill those criteria. > > Cheers > GL > > >> >> Regards, >> >> -Rob >> > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
