https://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&feed=atom&action=historyTalk:TransOgg Page - Revision history2024-03-29T12:44:53ZRevision history for this page on the wikiMediaWiki 1.40.1https://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12705&oldid=prevGmaxwell: /* Stream Alignment */2011-01-06T17:57:42Z<p><span dir="auto"><span class="autocomment">Stream Alignment</span></span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 10:57, 6 January 2011</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l36">Line 36:</td>
<td colspan="2" class="diff-lineno">Line 36:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Stream Alignment ==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Stream Alignment ==</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Several modern and upcoming file systems (e.g. BTRFS) support block level de-duplication. Partially duplicate files can share backing storage while remaining logically separate. For this to work the duplicated portions must have sufficient alignment— e.g. the FS can't discover duplicate blocks offset by 1 bytes. Perhaps we should consider some facility to insert a small amount of padding so that recovery points are aligned to some increment (it looks like 4k is sufficient). This way partially duplicate media files (e.g. editing intermediaries) could be effectively deduped. <del style="font-weight: bold; text-decoration: none;">The alignment art </del>recovery points would also have positive error robustness properties on block oriented media (e.g. if data is destroyed a sector at a time you want your coding units to cover a minimum number of partial sectors).--[[User:Gmaxwell|Gmaxwell]] 17:55, 6 January 2011 (UTC)</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Several modern and upcoming file systems (e.g. BTRFS) support block level de-duplication. Partially duplicate files can share backing storage while remaining logically separate. For this to work the duplicated portions must have sufficient alignment— e.g. the FS can't discover duplicate blocks offset by 1 bytes. Perhaps we should consider some facility to insert a small amount of padding so that recovery points are aligned to some increment (it looks like 4k is sufficient). This way partially duplicate media files (e.g. editing intermediaries) could be effectively deduped. <ins style="font-weight: bold; text-decoration: none;">Aligning </ins>recovery points would also have positive error robustness properties on block oriented media (e.g. if data is destroyed a sector at a time you want your coding units to cover a minimum number of partial sectors).--[[User:Gmaxwell|Gmaxwell]] 17:55, 6 January 2011 (UTC)</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12704:rev-12705 -->
</table>Gmaxwellhttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12704&oldid=prevGmaxwell: Add a section re: padding for block alignment.2011-01-06T17:56:35Z<p>Add a section re: padding for block alignment.</p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 10:56, 6 January 2011</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l34">Line 34:</td>
<td colspan="2" class="diff-lineno">Line 34:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Actually it is desirable, even if it is just for the stream ID. You'll win over no embedded programmers with the promise of rarely doing a lot of work rather than always doing a little, and you'll never guarantee that stream ID rewrites will be performed by all software when necessary when they're only necessary very infrequently. --[[User:Gumboot|Gumboot]] 09:14, 22 September 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Actually it is desirable, even if it is just for the stream ID. You'll win over no embedded programmers with the promise of rarely doing a lot of work rather than always doing a little, and you'll never guarantee that stream ID rewrites will be performed by all software when necessary when they're only necessary very infrequently. --[[User:Gumboot|Gumboot]] 09:14, 22 September 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>== Alignment ==</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>== <ins style="font-weight: bold; text-decoration: none;">Stream </ins>Alignment ==</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Several modern and upcoming file systems (e.g. BTRFS) support block level de-duplication. Partially duplicate files can share backing storage while remaining logically separate. For this to work the duplicated portions must have sufficient alignment— e.g. the FS can't discover duplicate blocks offset by 1 bytes. Perhaps we should consider some facility to insert a small amount of padding so that recovery points are aligned to some increment (it looks like 4k is sufficient). This way partially duplicate media files (e.g. editing intermediaries) could be effectively deduped. The alignment art recovery points would also have positive error robustness properties on block oriented media (e.g. if data is destroyed a sector at a time you want your coding units to cover a minimum number of partial sectors).--[[User:Gmaxwell|Gmaxwell]] 17:55, 6 January 2011 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Several modern and upcoming file systems (e.g. BTRFS) support block level de-duplication. Partially duplicate files can share backing storage while remaining logically separate. For this to work the duplicated portions must have sufficient alignment— e.g. the FS can't discover duplicate blocks offset by 1 bytes. Perhaps we should consider some facility to insert a small amount of padding so that recovery points are aligned to some increment (it looks like 4k is sufficient). This way partially duplicate media files (e.g. editing intermediaries) could be effectively deduped. The alignment art recovery points would also have positive error robustness properties on block oriented media (e.g. if data is destroyed a sector at a time you want your coding units to cover a minimum number of partial sectors).--[[User:Gmaxwell|Gmaxwell]] 17:55, 6 January 2011 (UTC)</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12703:rev-12704 -->
</table>Gmaxwellhttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12703&oldid=prevGmaxwell: /* Alignment */ new section2011-01-06T17:55:20Z<p><span dir="auto"><span class="autocomment">Alignment: </span> new section</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 10:55, 6 January 2011</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l33">Line 33:</td>
<td colspan="2" class="diff-lineno">Line 33:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::If it's still a desirable feature, then you can use a set of tables to edit each bit in the relevant field, or some collection of those tables representing the adjustment for various distances from the end of the checksum, and the only requirement would be that the possible distances can be constrained. --[[User:Gumboot|Gumboot]] 22:35, 4 June 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::If it's still a desirable feature, then you can use a set of tables to edit each bit in the relevant field, or some collection of those tables representing the adjustment for various distances from the end of the checksum, and the only requirement would be that the possible distances can be constrained. --[[User:Gumboot|Gumboot]] 22:35, 4 June 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Actually it is desirable, even if it is just for the stream ID. You'll win over no embedded programmers with the promise of rarely doing a lot of work rather than always doing a little, and you'll never guarantee that stream ID rewrites will be performed by all software when necessary when they're only necessary very infrequently. --[[User:Gumboot|Gumboot]] 09:14, 22 September 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Actually it is desirable, even if it is just for the stream ID. You'll win over no embedded programmers with the promise of rarely doing a lot of work rather than always doing a little, and you'll never guarantee that stream ID rewrites will be performed by all software when necessary when they're only necessary very infrequently. --[[User:Gumboot|Gumboot]] 09:14, 22 September 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">== Alignment ==</ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Several modern and upcoming file systems (e.g. BTRFS) support block level de-duplication. Partially duplicate files can share backing storage while remaining logically separate. For this to work the duplicated portions must have sufficient alignment— e.g. the FS can't discover duplicate blocks offset by 1 bytes. Perhaps we should consider some facility to insert a small amount of padding so that recovery points are aligned to some increment (it looks like 4k is sufficient). This way partially duplicate media files (e.g. editing intermediaries) could be effectively deduped. The alignment art recovery points would also have positive error robustness properties on block oriented media (e.g. if data is destroyed a sector at a time you want your coding units to cover a minimum number of partial sectors).--[[User:Gmaxwell|Gmaxwell]] 17:55, 6 January 2011 (UTC)</ins></div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12646:rev-12703 -->
</table>Gmaxwellhttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12646&oldid=prevGumboot: /* Checksum */ SIMD please.2010-10-17T10:59:48Z<p><span dir="auto"><span class="autocomment">Checksum: </span> SIMD please.</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 03:59, 17 October 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l17">Line 17:</td>
<td colspan="2" class="diff-lineno">Line 17:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial, to extend the principle to other media with other polynomials. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by layers closer to the media. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial, to extend the principle to other media with other polynomials. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by layers closer to the media. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">:: If there is a movement towards a non-CRC hash, please be careful to select something SIMDable. --[[User:Gumboot|Gumboot]] 10:59, 17 October 2010 (UTC)</ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12643:rev-12646 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12643&oldid=prevGumboot: /* Checksum */2010-10-17T09:29:47Z<p><span dir="auto"><span class="autocomment">Checksum</span></span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 02:29, 17 October 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l16">Line 16:</td>
<td colspan="2" class="diff-lineno">Line 16:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial <del style="font-weight: bold; text-decoration: none;">for extension onto </del>other media. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by layers closer to the media. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial<ins style="font-weight: bold; text-decoration: none;">, to extend the principle to </ins>other media <ins style="font-weight: bold; text-decoration: none;">with other polynomials</ins>. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by layers closer to the media. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12642:rev-12643 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12642&oldid=prevGumboot: /* Checksum */2010-10-17T09:10:15Z<p><span dir="auto"><span class="autocomment">Checksum</span></span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 02:10, 17 October 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l16">Line 16:</td>
<td colspan="2" class="diff-lineno">Line 16:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial for extension onto other media. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by the <del style="font-weight: bold; text-decoration: none;">medium</del>. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial for extension onto other media. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by <ins style="font-weight: bold; text-decoration: none;">layers closer to </ins>the <ins style="font-weight: bold; text-decoration: none;">media</ins>. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12641:rev-12642 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12641&oldid=prevGumboot: /* Checksum */ Avoid popular polynomials.2010-10-17T09:09:37Z<p><span dir="auto"><span class="autocomment">Checksum: </span> Avoid popular polynomials.</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 02:09, 17 October 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l16">Line 16:</td>
<td colspan="2" class="diff-lineno">Line 16:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>...though supporting multiple generator polynomials in a typical software implementation isn't hard.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>--[[User:Gmaxwell|Gmaxwell]] 06:56, 29 May 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">: If a message protected by a CRC fails to detect an error, then that message wrapped in header and footer and protected by the same CRC polynomial will fail to detect that same error. I think it should also be true that if that original message is split across several wrapper packets but the error appears entirely within one fragment, then the wrapper will still fail to detect the error. I think that's a strong case for avoiding 802.3's polynomial, but I'd suggest avoiding any popular polynomial for extension onto other media. I think a case could also be made for ignoring the low-BER assumption in choosing a polynomial, because that should already be handled by the medium. A pure hash is all that is needed here. In fact, do we just want a fast hash instead of a CRC? --[[User:Gumboot|Gumboot]] 09:09, 17 October 2010 (UTC)</ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Arrange checksum order for fast page edits of essential data ==</div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12477:rev-12641 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12477&oldid=prevGumboot: /* Arrange checksum order for fast page edits of essential data */ Prefer mandatory stream ID rewrites.2010-09-22T09:14:40Z<p><span dir="auto"><span class="autocomment">Arrange checksum order for fast page edits of essential data: </span> Prefer mandatory stream ID rewrites.</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 02:14, 22 September 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l30">Line 30:</td>
<td colspan="2" class="diff-lineno">Line 30:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Of course this is much easier for changing fields from one constant to another. If various bits change in different ways from page to page then you need a larger table of twiddles to mash into the CRC.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::Of course this is much easier for changing fields from one constant to another. If various bits change in different ways from page to page then you need a larger table of twiddles to mash into the CRC.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::If it's still a desirable feature, then you can use a set of tables to edit each bit in the relevant field, or some collection of those tables representing the adjustment for various distances from the end of the checksum, and the only requirement would be that the possible distances can be constrained. --[[User:Gumboot|Gumboot]] 22:35, 4 June 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>::If it's still a desirable feature, then you can use a set of tables to edit each bit in the relevant field, or some collection of those tables representing the adjustment for various distances from the end of the checksum, and the only requirement would be that the possible distances can be constrained. --[[User:Gumboot|Gumboot]] 22:35, 4 June 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">::Actually it is desirable, even if it is just for the stream ID. You'll win over no embedded programmers with the promise of rarely doing a lot of work rather than always doing a little, and you'll never guarantee that stream ID rewrites will be performed by all software when necessary when they're only necessary very infrequently. --[[User:Gumboot|Gumboot]] 09:14, 22 September 2010 (UTC)</ins></div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12226:rev-12477 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12226&oldid=prevGumboot: /* Arrange checksum order for fast page edits of essential data */2010-06-04T22:35:53Z<p><span dir="auto"><span class="autocomment">Arrange checksum order for fast page edits of essential data</span></span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 15:35, 4 June 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l27">Line 27:</td>
<td colspan="2" class="diff-lineno">Line 27:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Note that some CRC libraries and accelerators will have overheads that restrict their ability to accumulate data in a random order. They're likely to prefer well-aligned data and long runs of conventionally-ordered data. An ideal solution wouldn't interfere with their operation unnecessarily. --[[User:Gumboot|Gumboot]] 12:14, 4 June 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Note that some CRC libraries and accelerators will have overheads that restrict their ability to accumulate data in a random order. They're likely to prefer well-aligned data and long runs of conventionally-ordered data. An ideal solution wouldn't interfere with their operation unnecessarily. --[[User:Gumboot|Gumboot]] 12:14, 4 June 2010 (UTC)</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>:I guess we should figure out what fields people would want to change cheaply. The lacing, for example, isn't going to get changed without changing the data. I think the obvious cases include the DTS (among other things) and that's currently a variable length field. --[[User:Gmaxwell|Gmaxwell]] 15:06, 4 June 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>:I guess we should figure out what fields people would want to change cheaply. The lacing, for example, isn't going to get changed without changing the data. I think the obvious cases include the DTS (among other things) and that's currently a variable length field. --[[User:Gmaxwell|Gmaxwell]] 15:06, 4 June 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">::All I had in mind was the stream ID. In fact, perhaps making it easily editable would be preferable to making it large enough that it rarely needs to be changed.</ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">::Of course this is much easier for changing fields from one constant to another. If various bits change in different ways from page to page then you need a larger table of twiddles to mash into the CRC.</ins></div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">::If it's still a desirable feature, then you can use a set of tables to edit each bit in the relevant field, or some collection of those tables representing the adjustment for various distances from the end of the checksum, and the only requirement would be that the possible distances can be constrained. --[[User:Gumboot|Gumboot]] 22:35, 4 June 2010 (UTC)</ins></div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12225:rev-12226 -->
</table>Gumboothttps://wiki.xiph.org/index.php?title=Talk:TransOgg_Page&diff=12225&oldid=prevGmaxwell at 15:06, 4 June 20102010-06-04T15:06:13Z<p></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 08:06, 4 June 2010</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l26">Line 26:</td>
<td colspan="2" class="diff-lineno">Line 26:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br/></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Note that some CRC libraries and accelerators will have overheads that restrict their ability to accumulate data in a random order. They're likely to prefer well-aligned data and long runs of conventionally-ordered data. An ideal solution wouldn't interfere with their operation unnecessarily. --[[User:Gumboot|Gumboot]] 12:14, 4 June 2010 (UTC)</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Note that some CRC libraries and accelerators will have overheads that restrict their ability to accumulate data in a random order. They're likely to prefer well-aligned data and long runs of conventionally-ordered data. An ideal solution wouldn't interfere with their operation unnecessarily. --[[User:Gumboot|Gumboot]] 12:14, 4 June 2010 (UTC)</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">:I guess we should figure out what fields people would want to change cheaply. The lacing, for example, isn't going to get changed without changing the data. I think the obvious cases include the DTS (among other things) and that's currently a variable length field. --[[User:Gmaxwell|Gmaxwell]] 15:06, 4 June 2010 (UTC)</ins></div></td></tr>
<!-- diff cache key xiphwiki:diff::1.12:old-12224:rev-12225 -->
</table>Gmaxwell