On 27.05.2011 10:08, Rick Merrill wrote: --- Original Message ---
> Jay Garcia wrote: >> On 19.05.2011 18:03, Paul B. Gallagher wrote: >> >> --- Original Message --- >> >>> Jay Garcia wrote: >>> >>>> On 19.05.2011 15:56, Paul B. Gallagher wrote: >>>> >>>> --- Original Message --- >>>> >>>>> Jay Garcia wrote: >>>>> >>>>>> Since a prefetched page(s) is/are put to cache, I don't know if >>>>>> there is any indication that those pages are the ones >>>>>> "prefetched". >>>>>> >>>>>> And .. I think that this prefetch function is only workable as >>>>>> intended with a slow dialup type connection. My main website >>>>>> with over 800 pages is quickly accessed to any page from any page >>>>>> quite quickly. I don't really think I could tell the difference >>>>>> with/without prefetch. >>>>> >>>>> I like to watch videos online, and even with a broadband >>>>> connection (about 25 MB/min or 1500 kBps), they can sometimes take >>>>> awhile to load (probably due to slow/busy servers). For example, if >>>>> my 36-minute program is broken into three 12-minute chunks, I >>>>> routinely open three tabs, and launch all three chunks, then >>>>> quickly pause the second and third ones, allowing them to load >>>>> without playing. I view the first one, and by the time I'm done, >>>>> the second is ready to go. Automatic prefetching could come in >>>>> handy here, especially since the website links the second and third >>>>> chunks to the first and I have bandwidth to burn. >>>> >>>> Prefetch does nothing the first time you access the videos. And I >>>> don't think that's the purpose of prefetching anyway. Prefetch as I >>>> understand it is for pages with multiple links referencing other >>>> pages on the same site, >>> >>> ... which is exactly the case I'm describing. The page that embeds part >>> 1 of the vid has a link to part 2, and I could watch all of part 1 and >>> then click the link, or else I could right-click the link and say "open >>> in new tab." The second option is the one I choose -- I'm doing manual >>> prefetching, so that when I'm finished with part 1, part 2 is loaded and >>> ready to go. Some of the sites I visit even recommend this buffering >>> technique to avoid choppiness when a server can't keep up. >>> >>>> e.g., the prefetch link in the header of the index page would >>>> reflect<link rel="prefetch" ..>. >>> >>> ... and that's the answer to the OP's question: look in the source code >>> for<link rel="prefetch" ...>. If it has such a link, it's set up for >>> prefetching; if not, it's not. > > Has anyone seen or written such source code? This is the most logical > response, > and it presumably creates the desired result of making the remainder of > the site > faster to access. > >> You're not actually prefetching by definition, > > No. Getting data before it is needed IS "prefetching." > e.g. http://en.wiktionary.org/wiki/prefetch > >> just loading in another >> tab but accomplishing 'basically' the same thing. > > No. it is not another "tab". > e.g. http://en.wiktionary.org/wiki/prefetch > > In order to "fetch" something it has to be retrieved from somewhere in the local system. Prefetching in Windows fetches it's data from /Windows/prefetch. Firefox, SeaMonkey, etc. fetches it's data from cache. And that is the basis of my point simply because if you cache web sites, etc. it's the same as prefetching it when it's retrieved. I can't think offhand of prefetching any faster than from cache - memory or from disk. -- *Jay Garcia - Netscape Champion* www.ufaq.org Netscape - Firefox - SeaMonkey - Thunderbird _______________________________________________ support-seamonkey mailing list [email protected] https://lists.mozilla.org/listinfo/support-seamonkey
