dissenting.
The majority concludes that the solicitation was ambiguous and that this ambiguity was patent. I respectfully dissent from both of these conclusions. Instead, I find the solicitation unambiguous because only NVT’s interpretation thereof falls within a zone of reasonableness. In the alternative, I find any ambiguity to be latent. Thus, because NVT’s bid would have been lower than the government’s bid under the proper interpretation, I believe NVT should have been awarded the contract. I *1164would therefore reverse the decision of the Court of Federal Claims.
I
The majority’s conclusion that the solicitation was ambiguous is based on its finding that both NVT’s and the government’s interpretations were reasonable. While I agree as to NVT, I disagree with respect to the government. As a general matter, each of the hundreds of lines in the solicitation gave details for the listed task under columns titled “Number,” “Unit of Measure,” and “Frequency.” In addition, these bids were to be based on a one-year cost proposal, meaning that the “Frequency” column provides an “indication of how often, in a year, each of the listed tasks were to be performed.” Opinion at 1157. Accordingly, for each line item-including the twelve line items in question — NVT performed the same calculation, multiplying the value in the “Number” column by the value in the “Frequency” column to arrive at a yearly total. This, I believe, was reasonable for a number of reasons. To begin, NVT’s reading of the twelve disputed line items was consistent with its reading of the more than 800 other line items. In addition, NVT’s interpretation does not read out material parts of the solicitation, i.e., the column headings. NVT used the frequency values for these twelve line items just as it did for all the other line items; it did not ignore them.
Further, as the majority notes, the end result of NVT’s bid calculation method was also reasonable. The 287 identified buildings at the depot were built between 1918 and 1984, cover approximately 3,221,472 square feet of floor space on 433 acres, and are used by a population of up to 9,600 people. Thus, I agree with the majority that NVT reasonably interpreted the solicitation to require 76,209 square feet of tile replacement and 28,647.85 man-hours of labor. The estimated amount of tile replacement only amounted to a small percentage — two percent — of the total floor space. Opinion at 1161. This included 36,287 square feet of floor tile, approximately one percent of the total facility floor space, and 37,187 square feet of wall tile, plaster, or stucco. Moreover, these numbers also appear reasonable in light of the high maintenance standards stressed in the solicitation. Marine Corps policy dictated that “the unique buildings/facilities ... be maintained in a professional manner as a projection of Marine Corps pride that is world renowned.” To that effect, the Performance Work Statement required contractors to ensure that “all buildings, surfaces, and component parts remain in a serviceable condition and are free of warping, bucking, cracks, ... loose, missing or chipped tiles; noticeable surface imperfections, ... and other damage or defects.” Given the contract’s high standards, the majority correctly concluded that NVT’s bid did not call for a disproportionate amount of tile repair and was therefore reasonable.
I disagree, however, with the majority’s conclusion that the government’s interpretation of the solicitation was also reasonable (and therefore demonstrates the existence of an ambiguity). According to the majority, because “the numbers in the Frequency column are different from the words used elsewhere,” id. at 1161-62, and because they bear no relationship to any regular and periodic repetition, it was reasonable for the government to treat these twelve disputed line items differently. Specifically, in contrast to the more than 800 other line items that appear under identical column headings, the government did not use the value under the “Frequency” column as a multiplier for the value in the “Number” column. Id. In addition, the majority finds the government’s inter*1165pretation reasonable because the relevant estimate was consistent with the notion of incidental repair as well as proportional to what would be expected for a small set of line items that received no special treatment in the solicitation. Id.
I disagree for several reasons. First, given that the solicitation uses many different frequency indicators in a broad spectrum of formats, the fact that the disputed line items contained solely numbers did not justify the government’s (and the majority’s) dissimilar treatment thereof. This variety includes different primarily verbal, periodic indicators such as the following: every # years, Annually, # X Year, Semi-Annually, Quarterly, BiMonthly, Monthly, 2X Monthly, Bi-Weekly, Weekly, Daily, 2X Daily. But some line items give no specifics whatsoever, simply noting that the work needs to be performed “As needed” or “As-Req.” Moreover, the solicitation includes line items that express frequency in terms of the number of “treatments a year” and the number of “times a year.” In addition, the solicitation contains line items that give frequency in terms of hours and days, without any indication of periodicity. For example, with respect to hours, the solicitation contains line items calling for “20,-000 hours,” “6000 hrs,” and “2500 hrs.” With respect to days, the solicitation contains line items given in terms of “18 days,” “60 Days,” “90 Days,” and “180 days.” It even includes a line item with the “Frequency” given as “7/30 Days.” Finally, in the twelve disputed line items, the solicitation uses purely numerical figures in the “Frequency” column.
No matter what format the line item may have been in, the solicitation clearly required the multiplication of the “Number” by the “Frequency” to arrive at the estimated annual cost. The fact that the twelve disputed line items do not contain any words does not make them any different. As demonstrated above, the line items come in many different formats, including words, a combination of words and numerals, and solely numerals. Moreover, the line items list frequency in both periodic and non-periodic terms. Accordingly, the twelve disputed line items do not differ from any standard format used throughout the solicitation because the solicitation has no such clear standard format. As with all these other line items, the government should have simply applied the default presumption that the value in the “Frequency” column represents a yearly (per year) number because the bid price was to be based on the cost for one year’s worth of maintenance.
Second, as a legal matter, the government’s interpretation violates the canons of contract interpretation that require (a) consistency, and (b) giving meaning to each contract term. First, the government’s interpretation is unreasonable because it does not apply consistently to the entire solicitation. The solicitation called for the multiplication of the “Number” and the “Frequency” to arrive at the amount of work or materials required. For more than 800 line items, as the government acknowledges, bidders had to use this methodology to assemble their bid. But the government did not similarly treat the twelve disputed line items. In order to interpret the contract terms consistently, the government should have multiplied these “Frequency” figures by the “Number,” applying the per-year presumption. As a corollary to the consistency issue, if the government did not intend the numbers given in the “Frequency” column to be a frequency multiplier, it could have drafted the solicitation to indicate such repairs were to be done “Annually,” thereby requiring multiplication by one. This would have been consistent with the remainder of the solicitation, which has hun*1166dreds of “Annually” line items. By using a different value in the “Frequency” column for these twelve line items, however, the government clearly meant something different.
Further, by ignoring the “Frequency” value with respect to the disputed line items, the government also violated the canon of contract interpretation that disfavors rendering terms superfluous. If these numbers had no purpose, as government counsel suggested at oral argument, they should not have been listed in the “Frequency” column. Moreover, by ignoring these numbers, the government not only unreasonably ignored the values given in the individual line items, but also the column heading, which clearly indicates that the numbers themselves represent a “Frequency” to be used for purposes of assembling a bid.
In sum, the solicitation unambiguously required the multiplication of the “Number” and “Frequency” columns for each line item to arrive at a yearly figure for purposes of bid assembly, including the twelve disputed line items. As such, I find NVT’s interpretation of the solicitation the only reasonable interpretation, and, for the reasons given above, disagree with the majority that the government’s interpretation was also reasonable.
II
Even if I agreed with the majority that an ambiguity existed here, I do not find that ambiguity patent. The majority concludes that the ambiguity is patent because the line items are “expressed in a manner so different from hundreds of other line items, yielding totals disproportionate to the remainder of the solicitation.” Opinion at 1162. Yet in reaching this conclusion, the majority abandons the reasons why it concluded an ambiguity existed in the first place. In particular, the majority disregards, in my view, the reasons it found NVT’s interpretation reasonable, all of which point in the direction of finding a latent rather than patent ambiguity.
In my opinion, the ambiguity was latent. It was not, as the majority concludes, obvious, gross or glaring. As discussed above, the disputed line items do not differ dramatically from the other more than 800 line items; there are many other formats in which the solicitation lists “Frequency.” NVT likely entered these line items, as with the hundreds of others, into a spreadsheet program, multiplying them out to arrive at a bid price. A look at NVT’s proposal reveals that only two percent of the total floor space requires retiling every year. Yet the facility includes at least 287 buildings built between 1918 and 1984 on 433 acres with 3,221,472 square feet of floor space heavily used by up to 9,600 people at a time. Furthermore, the mandate in the solicitation requires that contractors maintain the facilities at the highest standards in order to project Marine Corps pride. Given this context, it is the government’s and not NVT’s estimates that seem disproportionately low, calling for replacement of only 1,354 square feet of tile. This represents only 0.04% of the total square footage of the facility. Assuming an ambiguity existed, therefore, NVT’s interpretation was far more reasonable, reflecting the fact that any defect in the solicitation was latent.
The government drafted the solicitation and should bear the burden of any mistakes contained therein. In this case, the solicitation unambiguously called for bidders to multiply the “Number” and “Frequency” columns for each line item. Even if there was an ambiguity, the law requires that it be resolved against the drafter. As *1167discussed, the patent ambiguity exception to this rule does not apply here. Because NVT’s bid would have been lower than the government’s bid had they been calculated using the same methodology, I respectfully dissent and would reverse the Court of Federal Claims decision granting the government’s motion for summary judgment.